1. No category

Generation and Application of Recombinant
Antibody Fragments for Prostate Cancer
Detection
A thesis submitted for the degree of Ph.D.
By
Conor Hayes B.Sc. (Hons)
Based on research carried out at
School of Biotechnology,
Dublin City University,
Dublin 9,
Ireland.
Under the supervision of Professor Richard O’Kennedy.
Declaration
I hereby certify that this material, which I now submit for assessment on the programme
of study leading to the award of Doctor of Philosophy is entirely my own work, that I
have exercised reasonable care to ensure that the work is original, and does not to the
best of my knowledge breach any law of copyright, and has not been taken from the
work of others save and to the extent that such work has been cited and acknowledged
within the text of my work.
Signed: _______________
ID No.:
Date: _________________
ii
Acknowledgements
iii
Table of contents
Declaration ........................................................................................................................ ii
Acknowledgements ..........................................................................................................iii
Abbreviations .................................................................................................................... x
Units ...............................................................................................................................xiii
Publications and Poster Presentations ............................................................................ xiv
1.0 Abstract .................................................................................................................... xvi
Chapter 1 ......................................................................................................................... 1
Introduction ...................................................................................................................... 1
1.1 Section overview ......................................................................................................... 2
1.2 History of PCa ............................................................................................................. 2
1.2.1 Background to PCa .................................................................................................. 3
1.2.2 Environmental factors .............................................................................................. 4
1.2.3 PSA ........................................................................................................................... 5
1.2.4 PSA and its isoforms .............................................................................................. 10
1.2.5 Other potential markers of PCa ............................................................................. 14
1.3 Introduction to antibodies and immunity .................................................................. 16
1.3.1 Antibodies and their structure................................................................................ 16
1.3.2 Antibody classes ..................................................................................................... 17
1.3.3 Innate and acquired Immunity ............................................................................... 18
1.4 Antibody Production ................................................................................................. 20
1.4.1 Hybridoma technology ........................................................................................... 20
1.4.2 Recombinant antibodies ......................................................................................... 21
1.4.3 Phage display ......................................................................................................... 22
1.5 Immunoassays for PCa detection .............................................................................. 28
1.5.1 Current immunoassays for PSA detection ............................................................. 29
1.5.2 History of PSA immunoassay development for clinical use ................................... 29
1.6 PSA assays for forensic applications ........................................................................ 32
1.7 Novel immunoassay formats reported in the literature ............................................. 34
1.8 Biosensor devices ...................................................................................................... 36
1.9 Thesis Outline ........................................................................................................... 41
Chapter 2 ....................................................................................................................... 42
Materials & Methods...................................................................................................... 42
iv
2.1 Materials & Methods................................................................................................. 43
2.1.1 Materials ................................................................................................................ 43
2.1.2 Culture media formulation ..................................................................................... 44
2.1.3 Buffer compositions................................................................................................ 44
2.1.4 Equipment list ........................................................................................................ 46
2.1.5 Bacterial strains used ............................................................................................. 47
2.2 Production of murine recombinant antibody fragments............................................ 48
2.2.1 Preparation of bacterial stocks .............................................................................. 48
2.2.2 Immunisation of Balb/c mice with fPSA and antibody titre determination ............ 48
2.2.3 Extraction and isolation of total RNA from the spleen of an immunised
Balb/c mouse ................................................................................................................... 49
2.2.4 Reverse transcription of total RNA to cDNA ......................................................... 49
2.2.5 PCR Primers .......................................................................................................... 50
2.2.6 Components and PCR conditions for antibody variable domain genes
amplification using Krebber-specific primers ................................................................ 52
2.2.7 Purification of VH and VL variable gene fragments using the Promega gel
clean up Kit ..................................................................................................................... 53
2.2.8 Splice by Overlap extension PCR (SOE-PCR)....................................................... 54
2.2.9 pAK 100 vector purification using Wizard miniprep DNA purification system .... 61
2.2.10 SOE-PCR restriction using SfiI enzyme and ligation into pAK 100 vector
for phage display analysis and ethanol precipitation ..................................................... 62
2.2.11 Preparation of electrocompetent E. coli cells ...................................................... 63
2.2.12 Electro-transformation of XL-1 blue E. coli cells with an scFv-containing
plasmid ............................................................................................................................ 64
2.2.13 Rescue of scFv-displaying phages and precipitation of phage-displayed
fragments ......................................................................................................................... 65
2.2.14 Phage precipitation using PEG/NaCl .................................................................. 65
2.2.15 Selection of antigen-binding phage scFv fragments by biopanning using
immobilised antigens....................................................................................................... 66
2.2.16 Polyclonal and monoclonal phage ELISA and scFv check. ................................. 68
2.2.17 Inhibition phage ELISA ........................................................................................ 70
2.2.18 ScAb expression vectors (pMoPAC 16 & 53), plasmid preparation and gel
purification ...................................................................................................................... 71
v
2.2.19 Vector and SOE-PCR product digestion .............................................................. 71
2.2.20 Sub-cloning the restricted scFv fragment into the restricted vector
fragments for scAb expression ........................................................................................ 73
2.2.21 Soluble expression of scAb fragments in pMoPAC 16 and pMoPAC 53
vectors ............................................................................................................................. 74
2.3 Generation of chicken-derived anti-fPSA recombinant antibody library ................. 75
2.3.1 Removal of spleens from chickens immunised with fPSA, RNA extraction and
cDNA synthesis ............................................................................................................... 75
2.3.2 PCR Primers .......................................................................................................... 75
2.3.3 Amplification of VH and VL chain genes for Fab library generation ..................... 77
2.3.4 Transformation into XL-1 Blue E. coli cells .......................................................... 80
2.4 Murine anti-fPSA scAb analysis using the Biacore 3000 instrument ....................... 82
2.4.1 Mouse monoclonal anti-human kappa light chain purification ............................. 82
2.4.2 Biacore maintenance .............................................................................................. 83
2.4.3 Preconditioning of the CM5 dextran sensor chip .................................................. 83
2.4.4 Preconcentration studies........................................................................................ 84
2.4.5 Immobilisation conditions ...................................................................................... 84
2.4.6 High throughput analysis of murine anti-fPSA scAbs using Biacore 3000 ........... 85
2.4.7 Kinetic analysis of selected clones using Biacore 3000 instrument ...................... 86
2.5 Murine anti-fPSA scAb clone sequencing ................................................................ 86
2.6 Selection of most sensitive murine anti-fPSA scAb by ELISA ................................ 86
2.7 Large-scale expression of a murine anti-fPSA scAb and subsequent
purification using immobilised metal affinity chromatography (IMAC) ....................... 87
2.8 Induced scAb cell expression profile ........................................................................ 88
2.9 Sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) .......... 88
2.10 Western blotting ...................................................................................................... 89
2.11 Avian clone analysis on Biacore A100 ................................................................... 89
2.12 Epitope Mapping of murine and avian anti-fPSA recombinant antibody
fragments using the Biacore 3000 instrument ................................................................. 90
2.13 Biacore subtractive inhibition assay for the detection of fPSA using the
purified murine anti-fPSA scAb ..................................................................................... 91
2.14 Murine anti-fPSA scAb inhibition ELISA .............................................................. 92
vi
2.15 Biacore sandwich assay using commercial anti-tPSA mAb (CanAg PSA 36
mAb) ............................................................................................................................... 92
2.16 Biotinylation of purified murine anti-fPSA scAb for sandwich assay
development .................................................................................................................... 93
2.17 ELISA-based assay for detection of biotinylated murine anti-fPSA scAb B5 ....... 96
2.18 Sandwich assay development for improved assay sensitivity................................. 96
2.19 Biacore sandwich assay using a neutrAvidin-immobilised CM5 dextran
surface ............................................................................................................................. 97
2.20 ELISA sandwich assay using a neutravidin-coated ELISA plate ........................... 97
2.21 Neutravidin ELISA plate coating concentration optimisation for improved
sandwich assay performance ........................................................................................... 98
2.22 TMB incubation time optimisation ......................................................................... 98
2.23 Inter-day and intra-day studies on sandwich assay for precision and accuracy
determination................................................................................................................... 98
2.24 Determination of the limit of detection (LOD) for fPSA in samples ...................... 99
2.25 Sandwich assay in spiked female serum ................................................................. 99
2.26 Patient serum sample analysis ................................................................................. 99
2.27 Immunohistochemical analysis ............................................................................. 100
2.27.1 Sample preparation for murine and avian anti-fPSA analysis (by hand) .......... 100
2.27.2 Immunohistochemistry using avian and murine anti-fPSA recombinant
antibody fragments ........................................................................................................ 100
2.27.3 Protocol for immunohistochemistry for diagnostic antibody to PSA (Dako)
(automated immunohistochemistry) .............................................................................. 101
Chapter 3 ..................................................................................................................... 102
The generation of recombinant antibodies to human prostate-specific antigen
using avian and murine immune systems ................................................................... 102
3.1 Introduction ............................................................................................................. 103
3.2 Murine antibody library generation ........................................................................ 104
3.2.1 Mice immunisation with fPSA antigen ................................................................. 104
3.2.2 Primer selection for murine heavy and light chain gene amplification ............... 105
3.3 Isolation of RNA from murine B cells (spleen) and first-strand DNA synthesis ... 106
3.4 PCR buffer optimisation for murine variable heavy and light chain generation .... 108
3.5 Murine SOE-PCR of variable heavy and light chains ............................................ 109
vii
3.6 Selection of fPSA-specific phage-scFv particles by ‗biopanning‘ ......................... 114
3.7 Murine polyclonal phage ELISA ............................................................................ 115
3.8 Murine monoclonal phage ELISA .......................................................................... 116
3.9 Soluble expression of mouse anti-fPSA scAb fragments ....................................... 119
3.10 Differential expression profile of an anti-PSA scFv and reformatted scAb ......... 121
3.11 Selection of chicken anti-fPSA scFv fragments.................................................... 122
3.12 Optimisation of avian VH and VL chain genes amplification for Fab library
construction ................................................................................................................... 123
3.13 Chicken anti-fPSA scFv selection on fPSA-coated ELISA wells ........................ 126
3.14 Soluble expression of chicken anti-fPSA scFv and analysis by ELISA ............... 129
3.15 Discussion ............................................................................................................. 132
Chapter 4 ..................................................................................................................... 135
Screening and characterisation of putative anti-fPSA scFv and Fab antibodies ..... 135
4.1 Introduction ............................................................................................................. 136
4.2 Biacore capture assay design .................................................................................. 137
4.2.1 Preconcentration studies of anti-human C kappa mAb ....................................... 137
4.2.2 Immobilisation of anti-human C kappa mAb on to the CM5 dextran sensor
chip surface ................................................................................................................... 137
4.2.3 Designed custom kinetic wizard ........................................................................... 142
4.2.4 Alignment of scAb clones from murine anti-fPSA library.................................... 148
4.2.5 Protein expression, extraction, purification and analysis ................................... 150
4.2.6 Time course expression profile analysis for scAb expression .............................. 153
4.3 Avian clone analysis ............................................................................................... 156
4.3.1 Analysis of the avian anti-fPSA Fab and scFv fragments on the Biacore
A100 instrument ............................................................................................................ 156
4.4 Determination of antibody pairs for sandwich assay development ........................ 165
4.5 Discussion ............................................................................................................... 167
Chapter 5 ..................................................................................................................... 172
Assay development and immunohistochemistry studies for fPSA detection .............. 172
5.1 Introduction ............................................................................................................. 173
5.2 Development of a subtractive inhibition Biacore assay for fPSA ........................... 175
5.3 Murine anti-fPSA inhibition ELISA assay with inter-day and intra-day studies .... 175
5.4 Biotinylation of the purified murine anti-fPSA scAb and assay development ........ 176
viii
5.5 Discussion ............................................................................................................... 188
Chapter 6 ..................................................................................................................... 192
Overall Conclusions ..................................................................................................... 192
6.1 Overall conclusions ................................................................................................. 193
Chapter 7 ..................................................................................................................... 196
Bibliography ................................................................................................................. 196
ix
Abbreviations
Ab – antibody
Abs – absorbance
AP – alkaline phosphatase
bp – base pair
BSA – bovine serum albumin
BPH – benign prostatic hyperplasia
cDNA – complementary deoxyribonucleic acid
CDR – complementary determining region
CH1 – constant heavy chain 1
CH2 – constant heavy chain 2
CH3 – constant heavy chain 3
CL – constant light chain
CM5 – carboxymethylated dextran
cfu – colony forming units
DMSO – Dimethyl sulfoxide
DNA – deoxyribonucleic acid
dNTP – Deoxyribonucleic triphosphate
E. coli – Escherichia coli
EDC – N-ethyl-N'-(dimethylamiopropyl) carbodiimide
EDTA – ethylenediaminetetra-acetic acid
ELISA – enzyme-linked immunosorbent assay
Fv – variable fragment
Fab – antibody binding fragment
FDA – Food and Drug Administration
FCA – Freund‘s Complete Adjuvant
FR – framework region
fPSA – free prostate-specific antigen
HA – haemagglutinin
HBS – HEPES buffered saline
HV – hypervariable
hK11 – human kallikrein 11
x
HRP – horseradish peroxidase
I – iodine
Ig – immunoglobulin
IMAC – immobilised metal affinity chromatography
IMS – industrial methylated spirits
IPTG – isopropyl-β-D-galactopyranoside
LAST – Laboratory animal science and training
mAb – monoclonal antibody
MOPS – (3 (N-Morpholino) propanesulfonic acid
mRNA – messenger RNA
MW – molecular weight
NHS – N-hydroxysuccinimide
NTA – nitrilotriacetic acid
OD – optical density
PAGE – polyacrylamide gel electrophoresis
PBS – phosphate buffered saline
PBST – phosphate buffered saline tween
Pan - biopanning
PCa – prostate cancer
PCR – Polymerase chain reaction
PEG – polyethylene glycol
POCT – point-of-care testing
pNPP – p-Nitrophenyl phosphate
PSA – prostate-specific antigen
RNA – ribonucleic acid
RI – refractive index
RPM – revolutions per minute
scAb – single chain antibody
scFv – single chain variable fragment
SDS – sodium dodecyl sulphate
SEER – Surveillance Epidemiology and End Results
SERS – surface enhanced raman scattering
SOC – super optimal catabolities
xi
SOE – splice by overlap extension
SPFS – surface plasmon field-enhanced fluorescence spectroscopy
SPR – surface plasmon resonance
tPSA – total prostate-specific antigen
TMB – Tetramethylbenzidine dihydrochloride
TYE – Tryptone yeast extract medium
T – Tween
VH – variable heavy chain
VL – variable light chain
xii
Units
µg – microgram
(k)Da – (kilo) Dalton
µL – microlitre
o
C – degrees Celcius
cm – centimetre
g – gram
h –hour
kg – kilogram
L – litre
m – metre
M – molar
mg – milligram
min – minute
mL – millilitre
mm – millimetre
nM – nanomolar
pg – picogram
rpm – revolutions per minute
RU – response unit
sec – second or s
v/v – volume per unit volume
x ‗g‘ – centrifugal acceleration
w/v – weight per unit volume
xiii
Publications and Poster Presentations
(i) A review ‗Biosensor developments: application to prostate-specific antigen
detection‘. D. A. Healy, C. J. Hayes, P. Leonard, L. McKenna and R. O‘Kennedy
(2007). Trends in Biotech., 3:125-131.
(ii) Leonard, P., Säfsten, P., Hearty, S., Butler, J., Jacobson, K. Hayes, C., and
O‘Kennedy, R. (2006). High throughput screening of recombinant antibodies. Invited
oral presentation at the Biacore UK Protein Interaction User Days, September 26th -27th,
Hinxton Hall, Cambridge, UK.
(iii) Finlay, W.J.J., Hayes, C., Leonard, P. and O‘Kennedy, R. (2006). Cloning and
analysis of specific Fab and scFv fragments for the recognition of diagnostic
biomarkers. Cambridge Healthcare Insititute‘s Third Annual Monoclonal Antibodies
Conference, Protein Engineering Summit, Boston, M.A., USA, April 24-28th.
(iv) Hayes, C., Leonard, P., Le Berre, M., Hayhurst, A., McKenna, L. and O'Kennedy,
R. (2007). Engineering Antibody Fragments for Forensic Applications. Accepted poster
presentation at the 6th International Congress on Recombinant Antibodies, June 26th28th, Hotel Palace Berlin, Germany.
(v) Hayes, C., Leonard, P., Le Berre, M., Hayhurst, A., McKenna, L. and O'Kennedy,
R. (2007). Engineering Antibody Fragments for Detection of Prostate Specific Antigen
for Forensic Applications. Accepted poster presentation at the Irish Society for
Immunology. September 13th-14th. School of Nursing, Dublin City University,
Glasnevin, Dublin, Ireland.
(vi) Hayes, C., Leonard, P., Le Berre, M., Hayhurst, A., McKenna, L. and O'Kennedy,
R. (2007). Engineering Prostate Specific Antigen Antibody Fragments for Forensic
Applications. Accepted poster presentation at the Europtrode IX Conference. March
30th-April 2nd. The Crowne Plaza, Santry, Dublin, Ireland.
(vii) R. O‘Kennedy, C. Hayes, P. Leonard. Antibodies and Sensors: Technologies and
Future Vision for Prostate Cancer Detection. Oral presentation (2007). Prostate Cancer
Research Consortium Research Meeting, April 27th. St. James Hospital, Dublin, Ireland.
(viii) R. O‘Kennedy, S. Hearty, P. Leonard, B. Byrne, A.D. Sheehan, C. Viguier, N.
Gilmartin, E. Tully, S. Stapleton, S. Townsend, C. Hayes, B. McDonnell, G. Donohoe,
M. Ryan, E. Darcy, and J. Fitzgerald Antibody engineering. (2006). US-Ireland R&D
xiv
Partnership Sensors Workshop, February 20th-21st. National Centre for Sensor Research
(NCSR), Dublin City University, Glasnevin, Dublin, Ireland.
xv
1.0 Abstract
Prostate cancer (PCa) is the most prevalent adenocarcinoma and the second highest
cause of cancer-related deaths in men. Early diagnosis is required to identify the
development of PCa to reduce the risk of the disease metastasising to different regions
of the body. Multiple biomarkers in serum have been identified for the diagnosis of
PCa. However, prostate-specific antigen (PSA) and its isoforms remain the gold
standard test for PCa detection. Many antibodies have been generated for the detection
of PSA with good sensitivities and limits of detection. To ensure reliable assay
performance a number of biomarkers for PCa are required to increase specificity and
sensitivity thus reducing the number of false positive results upon initial examination.
The overall aims of the work presented in this thesis were to generate, select and
characterise antibodies to PSA, an important biomarker in PCa.
The generation and selection of murine and avian recombinant antibody fragments to
free PSA (fPSA) was performed. Animals were immunised, RNA extracted and cDNA
synthesised. The antibody variable genes were amplified by PCR, gel-purified and
cloned into a phagemid vector. Electrocompetent cells were transformed and library
sizes of 2.02 x 106 cfu/mL and 3.10 x 108 cfu/mL were obtained for the murine and
avian clones, respectively. Specific clones were isolated by ‗biopanning‘ which
involved a selection process on fPSA-coated immunotubes, enrichment and affinity
maturation. Five and three rounds of selection were performed on the murine and avian
anti-fPSA single chain variable fragments (scFv) recombinant antibody libraries,
respectively. Preliminary analysis by ELISA indicated a large panel of positive binding
clones for both screened libraries with approximately 60% of the analysed clones
binding to fPSA. The murine recombinant antibody fragments were reformatted and
expressed as single chain antibodies (scAb). This construct greatly enhanced soluble
protein expression and stability while also facilitating the rapid kinetic analysis of
candidate antibodies by Biacore. A differential expression profile of an anti-fPSA scFv
and reformatted scAb was performed by ELISA.
Individual antibodies isolated from the ‗biopanning‘ process were then characterised.
The murine anti-fPSA antibodies were further analysed using the Biacore 3000 in a
xvi
capture assay format. This facilitates correct affinity ranking of the selected antibodies
where the antibody fragment was captured using the appropriate secondary tag antibody
(anti-human constant kappa Ab) and a fixed concentration of fPSA (1 g/mL) passed
over the captured antibody fragment. Ninety four antibody fragments were evaluated
and ranked on their percentage complex stability values after a 5 minute dissociation
period. Eleven of the ninety four were subjected to further analysis and their affinities
ranged from 3.99 x 10-9 M to 3.62 x 10-10 M. They were then expressed in host
Escherichia coli (E. coli) cells on a large scale and purified by immobilised metal
affinity chromatography (IMAC). The cell lysates were characterised by ELISA, SDSPAGE and Western blotting. The antibodies were sequenced and analysed for
variability and similarities between the coding regions of the variable domains.
Inhibition ELISA-based assays with inter and intra-day analysis were performed on the
purified antibodies and a reproducible and accurate assay format was demonstrated with
a limit of detection of 2.6 ng/mL. The avian clones were analysed, ranked and affinities
evaluated by Dr. Paul Leonard using the Biacore A100 employing the same assay
format used for analysis of the murine antibody fragments. Imperative to assay design is
the selection of optimal antibody candidates for improved performance. For assay
development antibody pairs which bind different regions of the fPSA molecule were
determined using the capture assay format on the Biacore 3000. One murine scAb, two
avian Fabs and one avian scFv were shown to bind different epitopes of the fPSA
molecule with no non-specific binding evident. Following the successful purification of
the murine and avian antibody fragments immunohistochemistry studies were
performed on PCa tissue samples. These antibodies were shown to bind to different
zones of the prostate gland (epithelial and stromal regions) confirming that the
antibodies were binding to different epitopes of the fPSA molecule.
The purified murine anti-fPSA scAb was biotinylated and this was employed in a
sandwich ELISA assay whereby the murine anti-fPSA biotinylated scAb was captured
on a neutravidin-coated ELISA plate and fPSA sandwiched using the avian antibody
fragments as the detection antibodies. The assay sensitivity was improved using this
format with a limit of detection of 1.8 ng/mL in PBST solution. Female serum was
spiked with fPSA and inter-day studies performed by sandwich ELISA. Patient serum
samples with diagnosed PCa were evaluated by sandwich ELISA to determine levels of
xvii
fPSA in the sample and data were compared to documented tPSA levels which were
comparable to the levels obtained from a commercially available sandwich assay.
xviii
Chapter 1
Introduction
1
1.1 Section overview
The aim of this thesis was to develop a sensitive immunoassay for the detection of
prostate cancer (PCa). This chapter describes the importance of early detection of PCa
and the impact of the disease on individuals. Potentially useful biomarkers for early
detection of PCa and current immunoassay tests available for their detection are
critically reviewed. A comprehensive description of antibody production and the
methods used for selecting antigen-specific antibodies is given. Sensitive detection
methods such as surface-plasmon resonance (SPR) and real-time polymerase chain
reaction (PCR) are also reviewed.
1.2 History of PCa
PCa was first discovered in 1853 (Lytton, 2001) and was initially regarded as a very
rare cancer. However, in the last 30 years an increase in our knowledge of cancer in
general and PCa in particular has occurred. Today 85% of new PCa cases are diagnosed
at local and regional stages and the 5 year relative PCa survival rate has increased by
20% since 1985 (Denmeade et al., 2004). The most widely used indicator to detect PCa
in the general population is a measurement of serum levels of prostate-specific antigen
(PSA), a serine protease produced by the prostate epithelium (Nadler et al., 1995). PSA,
the most widely used marker in clinical use for detection of PCa (Klotz and Teahan,
2006) became an FDA approved biomarker in 1987. Subsequent data demonstrating
that elevated levels of PSA could be used to detect PCa earlier than rectal examination
led to widespread PSA testing for early disease detection (Özena and Sözen, 2006).
Although raised levels of serum PSA are suggestive of PCa, diagnostic confirmation
requires a transrectal prostatic needle biopsy, a costly and invasive procedure requiring
prophylactic antibiotics (Roddam et al., 2005). Overall, survival rates have improved
when PCa is detected at an early stage but it can metastasize and spread to other regions
of the body. PCa is an indolent disease and many more men die with the disease than
die from the disease but its diagnosis severely impacts the patient‘s life and contributes
to anxiety and stress.
2
1.2.1 Background to PCa
Adenocarcinoma of the prostate is now the most frequently diagnosed male malignancy
with 1 in every 11 men developing the disease (Hayward, 2000). Each year in the UK
more than 25,000 men are diagnosed with PCa and approximately 10,000 die from the
disease (Zhu et al., 2006). Excluding skin cancers, PCa is the most common cancer in
men in the United States and the leading cause of cancer deaths in American men who
do not smoke (Özena and Sözen, 2006). In Europe, an estimated 2.6 million new cases
of cancer are diagnosed each year. Prostate cancer constitutes about 11% of all cancers
in Europe and accounts for 9% of all cancer deaths among the male population within
the EU (Aus et al., 2005). The incidence and mortality numbers world-wide are
approximately 679,000 and 221,000, respectively (Aus et al., 2005; Zhu et al., 2006).
PCa is classified as an adenocarcinoma, or glandular cancer of the prostate gland. The
prostate gland a walnut sized organ located in the front of the rectum and below the
bladder (http://www.psa-rising.com/prostatecancer/prostate.htm). Its principle function
is the production of seminal fluid that carries sperm. The majority of the seminal fluid is
produced by the seminal vesicles which are located directly behind the prostate gland
(http://www.mamashealth.com/organs/prostate.asp). The prostate encompasses part of
the urethra, the tube that expels urine from the bladder and semen through the penis.
Cancer occurs when the prostate cells mutate and start to multiply out of control. These
cells may metastasize from the prostate to other parts of the body, especially the bones
and lymph nodes causing pain, difficulty in urinating and other symptoms. Early PCa
usually produces no symptoms. Sometimes, however, PCa does cause symptoms, often
similar to those of non-malignant diseases such as benign prostatic hyperplasia (BPH).
These include frequent urination, increased urination at night, difficulty starting and
maintaining a steady stream of urine, blood in urine, and painful urination. PCa may
also cause problems with sexual function, such as difficulty achieving erection or
painful ejaculation (Miller et al., 2003).
3
1.2.2 Environmental factors
The specific causes of PCa remain unknown, yet several key factors have been
associated with an increasing risk of PCa. Age, diet, lifestyle, medications, race and
genetic makeup have long been recognised as contributors to PCa. The primary risk
factor is age. Prostate cancer is uncommon in men under 45, but becomes more
common with advancing age. The average age at the time of diagnosis is 70 (Hsing and
Chokkalingam, 2006). One of the most striking characteristics of PCa is the degree of
geographic variation in its patterns of occurrence and progression (Klassen and Platz,
2006). Higher rates are seen in the United States and Western Europe, with peak rates
among black men in the United States, while the lowest rates are observed in Asia
(Hsing et al., 2000). An interesting observation is the increase risk of PCa among
Asians when they emigrate to North America. Thus, the connection between
environmental and lifestyle-related factors is further strengthened (Whittemore et al.,
1995). The enhanced incidence of PCa is often linked with the high fat intake of
Western nations over those of Asia. In the U.S. the diet is rich in animal fats and meats
and poor in fruit and vegetables. Previous studies involving 51,529 men have shown
that increased total fat intake, animal fat intake, and consumption of red meat are linked
with an increased risk of PCa (Giovannucci et al., 1993). Dietary amounts of certain
foods such as vitamins and minerals can also contribute to PCa risk. Men with higher
serum levels of the short-chain -3 fatty acid linolenic acid have higher rates of PCa
(Gann and Giovannucci, 2005).
Other dietary factors that may increase PCa risk include low intake of vitamin E,
lycopene, omega-3 fatty acids, the mineral selenium and low blood levels of vitamin D
(Schulman et al., 2001). Clinical trials have provided evidence that men given tomato
sauce-based pasta dishes for three weeks before radical prostatectomy had increased
lycopene levels in the blood and the prostate resulting in decreased oxidative genomic
damage in leukocytes and prostate cells (Chen et al., 2001). Similar results were
observed where high plasma levels of the antioxidant carotenoid lycopene, resulting
from high tomato intake, were associated with a reduced risk of PCa (Gann et al.,
1999).
4
There are also some links between PCa and medications and medical conditions. Use of
anti-inflammatory medicines such as aspirin and chlolesterol-lowering drugs known as
the statins may also decrease PCa risk (Jacobs et al., 2005; Shannon et al., 2005).
However, recent conflicting studies have shown there may be an increased risk of PCa
development with statin use in obese men (Charnow, 2008). The use of statins in
patients with high cholesterol levels can mask subclinical PCa by lowering serum PSA
levels. Infection or inflammation of the prostate (prostatitis) may increase the chance for
developing PCa. In particular, infections with sexually transmitted infections syphilis,
chlamydia, and gonorrhea seem to increase PCa risk (Dennis et al., 2002).
Increased incidence of PCa found in certain racial groups, in identical twins with PCa
and in men with certain genes indicates genetic background contributes to the risk of
developing PCa. In the US, PCa affects black men more than white or Hispanic men
and is also more deadly in black men (Hoffman et al., 2001). Studies of twins in
Scandinavia suggest that forty percent of cancer risk can be explained by inherited
factors (Lichtenstein et al., 2000). While no single gene is responsible for PCa, two
genes (BRCA1 and BRCA2) are important risk factors for ovarian cancer and breast
cancer in women and have also been implicated in PCa (Struewing et al., 1997).
1.2.3 PSA
PSA (Figure 1.1) is produced mainly by the prostate gland and its function is thought to
be liquefaction of seminal fluid through enzymatic action. The active site of PSA is
closely related to that of chymotrypsin (Vihinen, 1994). PSA is not a cancer-specific
protein. Its presence is not unique to cancerous cells, as it is present in normal prostate
tissue (Jung et al., 2000a). Its use as a marker for PCa is due to the elevated levels
detectable in the serum of PCa patients.
5
Figure 1.1: The structure of horse prostate kallikrein (HPK) isolated from stallion
plasma as an example of a homologue of Human PSA (PDB accession codes 1PFA).
HPK was the first structure of a serine protease purified from seminal plasma and shares
extensive sequence homology with human PSA, including the predicted chymotrypsin-like
specificity (Carvalho et al., 2002). Image accessed using Protein Data Bank website
(http://www.rcsb.org/pdb/home/home.do). The catalytic triad residues (His57, Asp102
and Ser195) within the active site are shown.
Under normal conditions, i.e. in non-cancerous tissue, PSA synthesised by the epithelial
cells of the prostate gland is secreted into the seminal fluid. PSA can only enter the
bloodstream by leaking into the extracellular fluid and diffusing into veins and
capillaries. However, in cancerous prostatic tissue, tumour growth disrupts the basal cell
layer and cell polarity is altered (Figure 1.2). The direction of PSA secretion is
disturbed and PSA is released freely into the circulatory system.
6
Figure 1.2: The prostate is a walnut-sized gland located in front of the rectum and below the
bladder. PCa occurs when the cells mutate and begin to multiply out of control and the
structure orientation of the gland is breached. This diagram was obtained from
http://www.fitness-health.co.uk/prostate-cancer.htm.
This explains the elevated PSA levels in the serum of PCa patients (Stenman et al.,
1999). Even though PCa cells produce less PSA than non-cancerous cells, more leaks
into the serum and, therefore, elevated serum PSA is detected (Jung et al., 2000b). At
present, PSA remains one of the best studied PCa-associated markers. The discovery of
different forms of PSA and the differences in their relative ratios in PCa patients has
refocused research on PSA. However, the practical application of PSA in cancer
detection has proved problematic. One major factor contributing to this stems from the
fact that PSA is not a truly cancer-specific marker (Levesque et al., 1995).
Various other benign disorders have been shown to increase serum PSA levels, and the
potential for false positive results has diminished the reliability of PSA assays. One
major disorder capable of increasing the level of PSA in serum is benign prostatic
hyperplasia (BPH Anderson, 2002). BPH is a non-malignant growth of the prostate
7
gland, characterised by similar distortion of the basal cell layer of the prostate gland and
free release of PSA into blood circulation, as previously described for PCa. Current
commercially available assays can often fail to distinguish PCa from such benign
disorders and, as a result, much research is underway to enhance specificity of PSA
assays so as to eliminate the high percentage of false positives. The discovery of various
forms of PSA in serum, including the presence of complexes, has provided a new angle
on the protein‘s role and significance as a disease marker.
In forensics PSA is an acceptable marker in suspected rape criminal cases even in
individuals with vasectomises. The marker can be detected in vaginal swabs or
extracted from clothing fabric of cadavers. The presence of PSA in the sample can
implicate an individual in a specific crime and this may be confirmed by DNA analysis.
PSA is a 34 kilo Daltons (kDa) glycoprotein that consists of 237 amino acids, and is
found in the serum in several isoforms (Table 1.1). It is a member of the kallikrein
(KLK) family which can be divided into 2 groups, plasma and tissue kallikreins. The
role of plasma kallikrein in the body includes activation of blood clotting, kinin
generation and inflammation. The tissue kallikreins, otherwise known as glandular
kallikreins, were discovered initially in urine in 1928 by Frey and Kraut (Garatinni and
Shore, 1966). The gene family, located on chromosome 19q13.2-q13.4 (Mahabeer and
Bhoola, 2000), are a subgroup of the serine protease family of proteolytic enzymes.
Fifteen human tissue kallikrein members have been identified and these proteins are
intensely studied because of their association with cancers.
8
Table 1.1: Forms of PSA found in serum
PSA form
Description
Relevance to PSA
Reference
detection
Free PSA
Complete PSA protein
Levels of PSA increase
Kuriyama et al.
that has no
in the serum of prostate
(1980);
modifications
cancer patients
Stowell et al.
(1991);
Khosravi et al.
(1995); Gregorakis
et al. (2005).
Complexed PSA
PSA that has formed
Levels of complexed
Matsumoto et al.
complexes with various
PSA decrease in the
(1999); Matsumoto
serum proteins e.g. 1-
serum of prostate cancer
et al. (2000);
chymotrypsin and 2-
patients, and the ratio of
Stephan et al.
macroglobulin and is
free to complexed forms
(2007).
inactive
of PSA can discriminate
prostate cancer from
benign disorders
ProPSA
Precursor forms of PSA
In serum studies of men
Peter et al. (2001);
that contains 7, 5, 4 and
with PCa, proPSA was
Catalona et al.
2 extra amino acids at
shown to improve
(2003);
the start of the protein
specificity for PCa
Kumar et al.
detection.
(1997);
Mikolajczyk et al.
(2001).
Benign PSA
BPSA contains 2
BPSA concentrations
Linton et al.
(BPSA)
internal peptide bond
are elevated in benign
(2003);
cleavages at lysines 145
prostatic disease
Stangelberger et al.
and 182
(2007).
Intact non-native
A non-native form of
Serum inPSA levels
Nurmikko et al.
PSA
PSA that is not cleaved
decrease in PCa
(2000);
at lysines 145
Nurmikko et al.
(2001);
Djavan et al
(2004).
9
Kallikrein proteins are useful serum biomarkers for diagnosis and prognosis of cancer
(Yousef et al., 2003) and a number of these biomarkers have been linked with ovarian,
breast, prostate, testicular, cervical, pancreatic, colon cancers (Obiezu and Diamandis,
2005) and lymphoblastic leukemia. Many of the KLK genes are frequently expressed in
certain tissue samples and are secreted into the serum or seminal plasma. The kallikreins
share similar homology (40–80%) with the greatest degree of similarity noted between
KLK2 and PSA (Pakioures et al., 2007).
1.2.4 PSA and its isoforms
Pro PSA is released into the lumen of the prostate and, since it contains a 7 amino acid
pro-leader sequence, is also termed (-7)pPSA. (-7)pPSA is converted to mature PSA by
human kallikrein 2 (hk2) and human kallikrein 4 (hk4). Partial degradation of (-7)pPSA,
leading to truncation of the pro-leader sequence, gives rise to other forms of pPSA
including (-2), (-4) and (-5)pPSA. The predominant form of tPSA is an1chymotrypsin PSA complex (PSA-ACT), which appears to form 65 – 95% of the total
PSA content of human serum (Zhang et al., 1999; Özena and Sözen, 2006). Other PSAprotein complexes have been found in trace amounts, such as PSA 1-protease inhibitor
(PSA-API) and PSA 2-macroglobulin (PSA-AMG) which is inactive. Uncomplexed or
‗free‘ PSA (fPSA) accounts for most of the remaining PSA in serum which is 5–35% of
the total PSA content. Free PSA is composed of 3 distinct isoforms, pro PSA (pPSA)
isoforms, benign PSA (BPSA) and intact PSA (inPSA) – all of these forms are
significant in the diagnosis of patients with PCa (Mikolajczyk et al., 2004) (Figure 1.3).
10
Figure 1.3: ProPSA is activated to PSA by hK2 through cleavage of the 7 amino acid proleader sequence. PSA is mainly secreted directly in the lumen of the prostate gland where it
liquefies the seminal plasma, provides nutrients and protects the sperm from the acidic
environment of the vaginal secretions. However, gland abnormalities such as BPH and PCa
cause an elevation in the levels of detectable PSA in the circulatory system. PSA isoforms such
as fPSA and cPSA can be detected to determine for early diagnosis of the disease and the focus
has moved towards these isoforms for early detection of PCa. This diagram was obtained from
Lilja et al. (2008)
PSA testing has revolutionised survival rates among men over the past 20 years (Djavan
et al., 2004). According to Surveillance, Epidemiology and End Results (SEER) data,
the relative 5 year survival rates in PCa had reached almost 100% by 2000, compared to
67% and 75% during 1973–1976 and 1983–1985, respectively (Weir et al., 2003). A
serum total PSA (tPSA) measurement of 4 ng/mL is generally regarded as the threshold
above which a prostate biopsy is required. The specificity of this PSA cut-off level is
11
poor, however, especially below 10 ng/mL, and is associated with significant false
positive and negative results. It has been estimated that between 60–75% of men in the
4–10 ng/mL range (also termed the diagnostic grey zone) undergo unnecessary biopsy
(Catalona et al., 1991, 1994; Brawer et al., 2000; Finne et al., 2000; Andriole et al.,
2005) while a PSA threshold of 4 ng/mL misses between 20–40% of PCa positive
individuals (Mistry and Cable, 2003; Thompson et al., 2004). A PSA threshold of 2.5 or
3 ng/mL is being adopted by a number of urology clinics and screening studies (Labrie
et al., 1999; Krumholtz et al., 2002) as an indication for biopsy in order to maintain
high levels of PCa detection, but incurs the cost of a potentially higher proportion of
unnecessary biopsies. Despite its importance in significantly decreasing the number of
individuals diagnosed with clinically inoperative, non-organ confined disease,
alternative biomarkers are required to increase the specificity and reliability of testing,
especially at low tPSA serum concentrations (2–10 ng/mL). Increased tPSA levels have
been reported in both benign prostatic hyperplasia and prostatitis (Hessels et al., 2004),
and as such an increased serum tPSA level is, therefore, only a sensitive marker for
prostate gland irregularities (Stephen et al., 2003). Current research is focusing on the
use of different PSA isoforms to improve the discrimination between a non-malignant
and malignant tumour.
Although the majority of tPSA found in serum is PSA-ACT (65-95%), the use of this
protein complex as a diagnostic tool may be valuable in early detection of PCa,
especially in the 2.5–4.0 ng/mL tPSA range (Naya et al., 2005). However, it still
remains unclear whether this marker alone will be sufficient in the diagnosis of PCa.
PSA-ACT levels are higher in men with PCa than in healthy normal men. Since its
approval by the FDA as a test for monitoring PCa, almost every study performed on
PSA-ACT as a specific test marker has shown better diagnosis capabilities than tPSA at
tPSA serum levels above 4 ng/mL (Brawer et al., 2000; Jung et al., 2000; Mitchell et
al., 2001; Sozen et al., 2005). Investigations into the measurement of percentage (535%) fPSA as an improved and useful diagnostic tool in differentiating PCa from BPH
have also been performed, since it is known that a higher percentage of fPSA in serum
correlates with a lower risk of PCa. Improved discrimination between PCa and BPH in
the 2–10 ng/mL tPSA range can be achieved by measuring f/tPSA ratio (Catalona et al.,
1998; Woodrum et al., 1998). It has been reported that by measuring the f/tPSA ratio or
12
PSA-ACT levels, up to 50% and 35% of unnecessary biopsies could be avoided in the
tPSA ranges of 4–10 ng/mL and 2–4 ng/mL, respectively, while maintaining PCa
detection rates at about 80% (Roddam et al., 2005).
Evidence exists that the measurement of proenzyme forms of PSA may improve the
detection of PCa (Mikolaczyk et al., 2004). As mentioned earlier, Pro PSA (pPSA),
which is converted to active PSA through a cleavage process involving removal of a 7
amino acid sequence from the pro-leader sequence between arginine and isoleucine
(Peyromaure et al., 2005), may have an important role to play in PCa detection.
Truncated forms of pPSA including (-2) pPSA and (-4) pPSA are of particular interest
as they are more resistant to conversion to mature PSA. Scientific data using mass
spectrometry investigated levels of intact (-7) pPSA and truncated pPSA in cancer
serum and found the combined levels to be more than half of fPSA in the sample (Peter
et al., 2001). (-7) pPSA was the major form detected with (-5) pPSA, (-4) pPSA and (2) pPSA present to a lesser extent. In serum studies of men with PCa, pPSA has been
shown to improve specificity for PCa detection (Sokoll et al., 2003; Khan et al., 2003).
Immunostaining studies in cancerous and benign tissues showed that (-2) pPSA stained
stronger in the cancerous tissue extracts than in the benign tissue extracts, indicating its
link with cancer. (-2) pPSA represents 25–95% of fPSA in the serum of PCa patients
and only 6–19% of fPSA in patients without PCa. The percentage of fPSA that is
comprised of the pPSA forms increase in aggressive type cancers and this %pPSA could
considerably enhance the specificity for PCa detection in men with tPSA levels of 2 –
10 ng/mL, especially in low tPSA levels (< 4 ng/mL Mikolaczyk et al., 2004;
Peyromaure et al., 2005). Catalona and co-workers (Catalona et al., 2003) reported that
in the analysis of 1091 serum samples from patients enrolled in PCa screening studies,
measurement of %pPSA (ratio pPSA/fPSA) would have spared 19% of unnecessary
biopsies for patients with tPSA levels of 2-4 ng/mL, compared to 10% and 11%
biopsies spared by measurement of fPSA and PSA-ACT, respectively. At higher tPSA
levels (4-10 ng/mL), measurement of % pPSA would have spared 31% of unnecessary
biopsies compared to 20% for fPSA and 19% for PSA-ACT.
Other PSA isoforms may also have a role to play in the generation of more specific tests
that can discriminate between PCa and benign conditions, thereby leading to fewer
13
unnecessary biopsies. A distinct enzymatically-inactive form of PSA containing two
internal bond cleavages at Lys145 and Lys182 has been identified in BPH tissue (Zhang
et al., 1995; Chen et al., 1997) and is termed BPSA. BPSA has been shown to be
significantly higher in the serum of non-cancerous men with elevated PSA (Linton et
al., 2003) and relatively lower in cancerous prostate tissue (Mikolajczyk and
Rittenhouse, 2003). BPSA analysis in serum in conjunction with pPSA immunoassays
may provide additional discrimination between PCa and BPH (Mikolajczyk and
Rittenhouse, 2003). In addition to the different proPSA isoforms and BPSA, the
remaining fPSA serum appears to be composed of intact PSA, termed inPSA (Noldus et
al., 1997; Hilz et al., 1999). It appears that an increase in inPSA is most likely to be
associated with benign disease as inPSA levels seem to decrease in cancer (Mikolaczyk
et al., 2004). This form of fPSA has shown some promise as a marker for distinguishing
PCa from benign disease (Nurmikko et al., 2001).
1.2.5 Other potential markers of PCa
PSA and its derivatives are recognised as the best current indicators for the presence of
PCa. However, alternative markers that could be used to detect prostate cancer with
increased specificity and sensitivity are being investigated, including other members of
the human tissue kallikrein family of proteins. The expression of human glandular
kallikrein 2 (hk2), which is 78–80% identical to PSA, increases during the change from
benign to primary prostate cancer (Darson et al., 1999). It has been reported that hk2
can improve the discrimination of PCa from benign disease when measured in
combination with tPSA and fPSA (Kwiatkowski et al., 1998; Partin et al., 1999). Hk2
may also be an important marker for PCa aggressiveness (Haese et al., 2001) and for
predicting and staging organ-confined tumours (Hessels et al., 2004). Other kallikrein
family members have also been identified as promising new candidate markers for PCa
diagnosis (Yousef and Diamandis, 2001), including human kallikreins 4 (hk4) (Obiezu
et al., 2002), 11 (hk11) (Diamandis et al., 2002), 14 (hk14) (Yousef and Diamandis,
2001), and 15 (hk15) (Diamandis and Yousef, 2002). For example, a study by
Nakamura and co-workers showed that a combination of hk11-to-tPSA ratio with
%fPSA values gave a PCa detection specificity of 51.5%, compared with detection
specificities of 10–31% and 10–45% using tPSA and %fPSA measurements alone,
14
respectively (Nakamura et al., 2003). The future use of panels of these kallikreins in
combination with PSA and other markers may thus provide more specific assays for
PCa that reduce the prevalence of unnecessary biopsies.
The potential of several other possible prostate cancer markers to improve disease
detection have been investigated, and markers reported to be associated with aggressive
cancers include E-cadherin (ECAD Rubin et al., 2001), enhancer of Zeste homologue 2
(EZH2 Varambally et al., 2002), hepsin (Magee et al., 2001), -methylacyl-coenzyme
A racemase (AMACR Rubin et al., 2002) among others. For example, Rhodes and coworkers have demonstrated that the ratio of E-cadherin (ECAD) to enhancer of Zeste
homologue 2 (EZH2) was associated with prostate cancer recurrence after radical
prostatectomy in a statistically significantly manner and may be useful in defining a
cohort of high-risk patients (Rhodes et al., 2003). High levels of circulating insulin-like
growth factor (IGF)-1 and low IGF binding protein-3 (IGFBP-3) have been identified as
being associated with increased risk of prostate cancer, although the reliability of these
serum markers for use in the detection of prostate cancer has been questioned (Ismail et
al., 2003). Serum IGF-11 levels are lower in PCa patients than men with BPH, and it
has been suggested that IGF-11 may serve as a useful marker for detecting PCa in men
with high PSA values (Trojan et al., 2006).
Reports on the usefulness of urinary biomarkers for the detection of PCa have also been
published. For example, it has been reported that the ratio of urinary PSA to serum PSA
(U/S PSA), when combined with the free to total serum PSA ratio, may be of benefit in
detection of PCa (Irani et al., 2005). In addition, the urinary biomarker thymosin 15
has been linked to PCa malignancy and its future recurrence. It has been reported that
the specificity of PCa detection using a tPSA cut-off of 2.5 ng/mL can be improved
from 55.4% to 70.8% by incorporating thymosin 15 measurements (Hutchinson et al.,
2005a). Attempts have been made to increase the value of thymosin 15 as a potential
biomarker in PCa detection by developing a quantitative assay to detect this protein in
bodily fluids (Hutchinson et al., 2005b).
15
1.3 Introduction to antibodies and immunity
1.3.1 Antibodies and their structure
Antibodies are glycoproteins produced when a foreign entity elicits an immune response
in the body. The body‘s defence system generates antibodies upon exposure to
potentially hazardous pathogens which eliminate and destroy them. The basic structure
of an antibody is shown in Figure 1.4.
Figure 1.4: (A) Ribbon structure of the immunoglobulin G (IgG) molecule. The molecule
consists of two identical heavy (H) chains (approximately 50 kDa) and two identical copies of
16
a light chain (25 kDa). Each light chain is attached to a corresponding heavy chain by a
disulphide bond. Both heavy chains are covalently linked by interchain disulphide bridges
located at the hinge axis. The heavy chain is divided into 4 domains (1 x VH and 3 x CH
domains) and the light chain is divided into 2 domains (1 x VL and 1 x CL). The VH and VL
domains confer antigenic binding specificity. (B) The complementary determining regions
(CDR) regions which form the antigen binding site, antigen binding fragment (Fab), variable
fragment (Fv) and constant fragments (Fc) are highlighted.
Polyclonal antibodies are a mixture of immunoglobulins (Ig) that recognise different
epitopes of the antigen, while monoclonal antibodies bind a single epitope of the
antigen. Monoclonal antibodies (Figure 1.4) were first reported by Georges Kohler and
Cesar Milstein in 1975. Since then many advancements in this scientific field have
generated recombinant antibody fragments namely, the single chain variable fragments
(scFv), the fragment antigen binding (Fab) and the single chain antibody fragment
(scAb). These fragments retain antigenic specificity due to the complementary
determining region (CDR) of the antibody without comprising binding affinity.
1.3.2 Antibody classes
There are 5 major classes of antibodies, IgM, IgG, IgA, IgD and IgE. These
immunoglobulins differ in both their structure and immune function. IgMs are
pentameric in structure and are clinically significant because they are predominant in
early immune responses to most antigens. IgAs are polymeric and are the predominant
immunoglobulin in saliva, tears, nasal mucosa, prostatic fluid and many other bodily
fluids. IgD antibodies are monomeric and trace amounts are present in serum and are
found on the surface of human B lymphocytes. IgE antibodies are present in serum in a
monomeric form and represent only a small fraction of total antibodies in blood. They
are involved in the production and release of vasoactive mediators e.g. histamine and
other chemicals that cause an inflammatory reaction. In healthy adults, the fourpolypeptide chain IgG monomer constitutes approximately 75% of the total serum
immunoglobulins (Mayer, 2006). The basic structure of an antibody consists of four
polypeptide chains with two identical heavy (H) chains and two identical light (L)
chains connected by disulphide bonds. The structure varies depending on the isotype.
The amino sequence in the variable region of the antibody varies greatly among
17
different antibodies. It is composed of 110-130 amino acids which give the antibody its
specificity for binding to the antigen. The variable region is subdivided into
hypervariable (HV) and framework (FR) regions. Within the variable heavy (VH) and
variable light (VL) chains three HV regions exist (HV 1, 2 and 3). These HV regions are
collectively known as complementary determining regions (CDRs). The CDRs are
interspersed by Framework (FR) regions. The FRs provide a backbone structure for the
antibody, and can also have an influence on antigenic specificity (Maher, 2006). The
conformation of the VH and VL chain CDRs result in six HV loop structures which
structurally form antigen-binding surfaces or pockets (Figure 1.4).
1.3.3 Innate and acquired Immunity
Immunity is the protection of the body from infectious diseases which may cause harm
and sickness. The innate immune system is a non-specific genetically derived system
passed on from mother to offspring. The natural body defence system encompasses
surfaces barriers e.g. skin, which cannot be penetrated by most organisms unless an
opening is present. It also consists of mucosal immunity e.g. saliva, tears, nasal
secretions and lysozyme, an enzyme in perspiration that cause cell lysis in Gram
positive bacteria. Acquired immunity involves a cascade of events involving B and T
lymphocytes (Roitt et al., 2006). B cells lymphocytes (generically referred as B cells)
are derived from the stem cells of the bone marrow and produce antibodies. T
lymphocytes (also known as T cells) are produced in the bone marrow but processed in
the thymus and constitute the basis of cell-mediated immunity. In cell-mediated
immunity, macrophages engulf the antigens, process them internally and display part of
the antigens on their surface along with their own proteins. This stimulates T cells to
recognise the presented antigens. Helper T cells (CD4+) are involved in directing the
immune response. They secrete lymphokines that encourage cytotoxic T cells and B
cells to grow and divide, attract neutrophils and boost the ability of macrophages to
engulf and destroy microbes. Suppressor T cells inhibit the production of cytotoxic T
cells and memory T cells recognise and respond to a further infection of the same
pathogen. Immature B-lymphocytes are stimulated to maturity when an antigen binds to
their surface receptors and T helper cells are in close proximity.
18
Light chain
V1V2V3 Vn
Heavy chain
J1 J2 Jn
Ck/
V1 V2 Vn D1 D2 D
5
J1 J2 Jn
C
3
n
'
'
GermLine DNA
D-J
recombination
Somatic
5
3
'
'
5
3
'
'
recombination
VDJ
5
3
'
'
5
3
Transcriptio
'
'
nPrimary
5
3
'
'
recombination
transcript RNA
Splicing
5
3
mRNA
'
'
Translation
5'
3
'
Polypeptide
chain
Figure 1.5: Somatic recombination of the separate gene segments for the construction of the
variable genes. Light chain genes are constructed whereby a variable (V) and joining (J) gene
segment are combined to form a complete light-chain V-region gene. The constant (C) is
encoded on a separated gene locus and is attached to the variable region exon through splicing
of the light-chain RNA to remove the J and C introns. Heavy chain genes are constructed using
three gene segments, diversity (D), J and C. Firstly the D and J gene segments are joined
followed by the combination of the V gene segment to form a complete heavy-chain exon. The
heavy chain C-region genes are encoded by numerous exons which are spliced to the heavy
chain C-region during processing of the heavy-chain transcript. The light and heavy chain
genes are translated into large polypeptide chains that associate to form the antibody molecule.
The diagram was adapted from Janeway et al., (1999).
The capability of the body to generate an enormous diversity of antibodies accounts for
the remarkable adaptive nature of the immune system. It is because of the unique
genetic mechanisms governing antibody synthesis that the body is capable of producing
these highly specific molecules to foreign entities. This diversity is mainly confined to
19
the VH and VL chains. VL chain genes are composed of two loci, the variable (V) and
joining (J) loci, while the VH genes are composed of three loci, the V and J loci with an
additional diversity (D) locus. During B-cell maturation, these different regions
combine randomly to produce a huge range of diverse antibodies. Recombination of the
various VDJ sequences in the VH chain and VJ sequences in the VL chain results in the
formation of different conformations in the binding sites of the antibody (Storb et al.,
2001; Janeway et al., 1999; Figure 1.5).
1.4 Antibody Production
1.4.1 Hybridoma technology
Over a century ago Paul Ehrlich (19th Century) coined the phrase ‗magic bullets‘ to
explain how antibodies might target and interact with their respective antigens. Seventy
five years later, Georges Kohler and Cesar Milstein invented the technology for cloning
individual antibodies (Bartal and Hirshaut, 1987). Monoclonal antibodies specifically
recognize one epitope of the cognate antigen. They can be generated using hybridoma
technology which involves fusing lymphocytes from the spleen of an immunised mouse
with an immortal cancer cell line (myeloma). The myeloma cell lines are selected so
that they do not possess the capability to produce immunoglobulins unless fused to
lymphocytes. This fusion is performed using polyethylene glycol (PEG) which is a
polywax solution that enhance adjacent cell fusions and the exchange of nuclei (Little et
al., 2000). A mixed population of fused cells, unfused myelomas and lymphocytes
results and these are incubated for 7 days before screening. The fused hybridoma cells
are selected using HAT (Hypoxanthine, Aminopterin and Thymidine)-containing
medium.
Myeloma
cells
lack
the
HGPRT
(hypoxanthine-guanine
phosphoribosyltransfer) are enzyme and when the de novo synthesis of purine and
pyrimidines is blocked by HAT addition the cells die, while lymphocytes which do not
grow in this culture media eventually die. Therefore, in the presence of HATsupplemented media only the hybridoma cells will proliferate. Secreted antibodies in
conditioned media from each hybridoma are tested against the immunised antigen by
ELISA. Positive hybridomas are scaled up and subsequently cloned (by limiting
20
dilution) to select for a single monoclonal antibody-producing hybridoma cell. The
hybridoma cells can be stored indefinitely in liquid nitrogen.
1.4.2 Recombinant antibodies
The development of recombinant techniques has facilitated the generation of
specifically engineered antibody fragments that have been successfully employed in
immunoassays (Dillon et al., 2003). Recombinant antibody fragments are becoming
essential tools in research due to their intrinsic properties such as greater penetrability,
ability to maintain antigen recognition, small size and ease of production when
compared to the complete antibody. In medicine, biotechnology and therapeutics use of
recombinant antibody fragments expressed in E. coli cells is becoming increasingly
popular.
Recombinant antibody fragments can be generated by directly cloning the heavy and
light chain variable domains from a monoclonal antibody-producing hybridoma cell or
amplifying the variable genes from cDNA derived from mRNA from lymphocytes of an
immunised source. The single-chain variable fragment (scFv) consists of heavy and
light chain variable domains of the full immunoglobulin antibody joined using a flexible
glycine-serine linker (Figure 1.6). The small fragment (30 kDa) retains the specificity
and affinity of the parental antibody (150 kDa) despite the exclusion of the constant
domains. The antigen-binding fragment (Fab) has generally been seen as a more
functional and stable version of recombinant antibodies than the scFv. The Fab consists
of the variable heavy and light chain genes joined with the constant heavy and light
chain genes developing a more stable antibody construct. Functional stability studies by
Hernandez et al., (2006) demonstrated the overall improved stability of a Fab over scFv
in guanidinium hydrochloride. In 1 M guanidinium hydrochloride 80% activity for Fab
was observed compared to 50% activity for an scFv. The affinity of the Fab and scFv
was not drastically affected by the antibody format but the functional stability was
significantly improved using the Fab format suggesting that the constant regions of the
Fab greatly enhance stability. Disadvantages of using murine scFvs include poor
expression levels with low product solubility, cell culture arrest and lysis, misfolding of
expressed antibody and protein aggregation (Hayhurst, 2000). Hayhurst and co-workers
21
partially addressed these problems by generating a single chain antibody fragment
(scAb) by fusing a human constant kappa light chain constant domain to the scFv and,
in addition, significantly improved expression levels were achieved by incorporating a
downstream periplasmic Skp chaperone gene. The elevated expression levels of scAb
were observed to be 10-fold better than the scFv.
Figure 1.6: Genetically-derived recombinant antibodies which retain antigen specificity are
shown. The smallest of these recombinant antibody fragments is a scFv fragment consisting of
the VH and VL chain of the full IgG molecule bound by a glycine-serine linker. The Fab fragment
is similar in structure to the scFv fragment but is a more stable construct. A scAb fragment
consists of linking a human constant kappa domain to the scFv fragment. The CDR regions of
the antibody fragment are responsible for binding to the epitope on the antigen.
1.4.3 Phage display
The selection of antigen-specific recombinant antibody fragments involves a process
known phage display technology. This technology was first reported in 1985 by Smith
and since then phage display has been extensively and exhaustively implemented in
many fields of research, in particular for antibody generation. Five years later
McCafferty et al., (1990) demonstrated that an antibody scFv gene could be directly
incorporated into the phage genome. The selection of antigen-specific antibodies from
recombinant antibody libraries is the most successful use of phage display (Bradbury
and Marks, 2004). Phage display involves using a filamentous bacteriophage which is a
circular singly stranded DNA genome encapsulated within surface coat proteins.
22
Foreign DNA can be inserted into regions of the genome and expressed on the surface
as fusion coat proteins. Thus, there is a direct link between the genotype and phenotype.
Figure 1.7: The recombinant scFv is expressed on the surface of a bacteriophage and the
gene encoding the fusion protein within the virion. Each pIII protein can display 2 – 3 copies
of the recombinant antibody fragment. Each phage infects E. coli cells and multiple copies
are propagated and selected using ‘biopanning’. Antigen-specific recombinant antibody
fragments are selected from large recombinant antibody libraries.
Phage display libraries can be successfully generated using an immunised source or
developed synthetically. Naive libraries are generated from unimmunised human V
genes, shuffled human V genes or synthetic human V genes while immune libraries are
derived from V genes from an immunised source. Normally, cDNA is generated from
the mRNA extracted from the spleen or bone marrow of an immunised animal. The
heavy and light chain genes are amplified by polymerase chain reaction (PCR) and the
antibody fragment inserted into one of the encoding genes for surface coat proteins. The
phagemid DNA is transformed into competent E. coli cells and after expression the coat
protein fusions are integrated into new phage particles assembled in the bacterium. This
enables phage particles to be replicated as ssDNA in the presence of helper phage
(NEB). Helper phages provide the necessary viral tools for the initiation and packaging
of phagemid DNA into phage particles. Helper phages contain a defective origin of
replication thus limiting their own packaging ability to produce more helper phage
molecules. This characteristic is important as it reduces the ability of helper phage to
replicate resulting in dominant phagemid DNA production. The advantages of
recombinant technology for antibody production include the fact that a highly
comprehensive library of antibody fragments can be generated from which an antibody
23
with high specificity for the antigen can be selected (Hoogenboom et al., 1998). Phage
particles are harvested from the infected cells and positive clones with desired affinities
selected by a process known as ‗biopanning‘ (Figure 1.8). The selection process can be
achieved by immobilising the antigen on a solid surface, incubating the antibodycontaining phage particles, removing non-specific phage by washing, eluting bound
phage with a high salt or low pH solution and re-infection into bacteria for replication,
enrichment and further rounds of panning with more stringent selection conditions. The
support for antigen immobilisation must possess a low phage binding capacity. The
target antigen-coating concentration must be sufficient for selection of specific phage
for further rounds of enrichment (Smith, 1985).
Figure 1.8: Individual phage particles displaying the scFv – pIII fusion proteins are selected
by ‘biopanning’. This process involves incubating the phage-scFv particle in an immunotube
coated with the antigen and washing away non-specific phage particles. The antigen-specific
phage particles are eluted and re-infected into E. coli cells for propagation and analysis in
subsequent rounds of selection. The stringency is increased by decreasing the antigen
concentration and increasing the washing steps to select high affinity recombinant antibody
fragments.
24
The phage pools from each round of selection are analysed by ELISA to confirm
antigen-specific antibody fragment enrichment. Individual clones are analysed,
screened, sequenced and affinities determined. Methods for the selection of antigenspecific recombinant antibodies have evolved, of which phage display has emerged as
one of the most valuable. The key aspect of phage display is that the genotype of a
particular antibody is linked through packaging inside a virion particle to the phenotype
of a display antibody fused to a coat protein of the phage particle.
Although phage display libraries have been generated using the pVIII major coat phage
protein, the pIII minor coat protein of filamentous phage is more widely acceptable
(Barbas et al., 2001, Figure 1.7). The clear advantage of pIII over the other coat proteins
is its ability to package large insertions, its capability to display monovalent copies of
the phage-antibody particle and the availability of suitable vectors. The filamentous
bacteriophages are a group of viruses capable of infecting Gram negative bacteria. The
most widely used are the Ff class (f1, fd and M13) which have shown to be 98%
homologous. The Ff phages infect E. coli cells with the F conjugative pilus and do not
kill the host cells during infection. The phage is 6.5 nm in diameter and 930 nm in
length with a covalently closed single stranded genome of 6400 nucleotides coding for
11 different proteins (pI-pXI) present within the coat proteins. Five of the 11 proteins
(pIII, pVI, pVII, pVIII and pIX) make up the flexible protein cylinder, 2 (pII and pX)
are involved in single stranded DNA (ssDNA) replication and 3 are required for phage
particle assembly. The length of the cylinder is composed of 2700 molecules of the
major coat protein (pVIII) while the minor coat proteins (pVII and pIX) are found at
one end of the phage particle with 5 molecules of both gene products. There are
approximately 5 copies of the other minor coat proteins (pIII and pVI) at the other end
of the phage particle. The package signal is located at the pVII and pIX end of the phage
particle and this is the first region to be assembled (Figure 1.7).
The infection process is a multistep process involving the F pilus on the surface of the
E. coli cell. When the phage comes in contact with the pilus, it retracts drawing the
phage particle into the cell. Pilus formation is at its most abundance at mid-exponential
phase of growth and decreases with increasing growth profile. The outer membrane
bacterial proteins (TolQ, R and A) are needed for translocation of the filamentous phage
25
DNA into the cytoplasm and the translocation of the phage coat proteins into the
cytoplasmic membrane. The initiation of infection occurs upon binding of the tip of the
F pilus to the pIII phage protein. The pIII phage protein consists of 3 domains (N1, N2
and CT) separated by glycine-rich regions. Each domain is essential for bacterial
infection. The N1 and N2 domains are in close proximity and N2 is responsible for
binding to the F pilus. The N1 domain interacts with the bacterial membrane protein
(TolA-D3) and the phage is retracted towards the bacterium surface where the major
and minor phage capsid proteins are disassembled and the phage DNA is translocated
into the cytoplasm. The viral (+) phage ssDNA strand enters the cytoplasm and the
complementary (-) DNA strand is synthesised by bacterial enzymes. This replicative
form (RF) is the template for transcription and phage protein translation. The pII protein
cleaves the intergenic region of the positive strand of the RF which acts as a primer for
synthesis of a new viral strand via a ‗rolling-circle‘ replication method continuing until
200 copies are within the cell. This continues until pV approaches a crucial level where
it binds to newly synthesised viral ssDNA. The newly synthesised ssDNA-pV (800 nm
in length and 8 nm in diameter) is assembled in the cytoplasm of the bacterium and the
phage particle is subsequently secreted from the cell. This is then cleaved and used as a
template for further phage propagation. Approximately 1000 phage per cell are
produced in the first hour with 100 – 200 phage particles in subsequent cycles.
Consideration of a vector for the display of proteins or peptides as coat protein fusions
to pVIII or pIII is important. In phage vectors the gene is inserted directly into the
genome and expressed with multiple copies of displayed protein leading to polyvalent
display. This causes the selection of lower affinity variants during the biopanning
process. To avoid an avidity affect a monovalent phage display process was developed
using a phage-plasmid vector (phagemid). This is a plasmid with a plasmid origin of
replication and a phage-derived origin of replication. The vector is capable of producing
large quantities of pIII-fusion proteins but cannot make phage. This is aided by the
addition of helper phage (i.e. M13K07 and VCSM13) that supply all the phage enzymes
and proteins needed for phage replication (Barbas et al., 2001). As mentioned
previously the phagemid genomes can be packaged in the phage coat preferentially
rather than in helper phage as these contain a defective origin of replication, which thus
ensure an abundance of pIII fusion protein. In phage assembly only 10% of the
26
packaged phage display the pIII fusion protein with 1–2 fusion proteins displayed on the
phage. The low percentage is due to the preferential assembling and packaging of wild
type pIII into phage. In contrast to phage vectors, phagemid vectors offer both a plasmid
origin of replication and a phage-derived origin of replication with a selectable
antibiotic resistance gene for facile selection and propagation of pIII fusion proteins.
The emergence of tetracycline resistance in cells not infected with phage was observed.
Thus, cultures for phage display should be grown fresh from a reliable frozen stock. The
phagemid vector also contains a lac promoter which controls expression levels with
ompA and pelB leader sequences directing the protein expression to the bacterial
periplasm.
Soluble expression of recombinant antibody fragments can be achieved by changing the
bacterial strain. An amber stop codon (TAG) is present between the protein gene and
the phage gene. During the selection process the phagemid is expressed in a suppressor
E. coli strain (XL1-Blue, TG1) which allows for transcription of the pIII fusion protein.
By incorporating the phage into a non-suppressor E. coli strain (TOP10F) the amber
stop codon is read with soluble expression of the recombinant antibody fragment in the
periplasmic space without the pIII gene product. Recombinant antibodies can be
expressed with fusion tags (hemagglutinin and His) for downstream detection and
purification.
Recombinant antibodies can be genetically manipulated by techniques such as affinity
maturation (Schier et al., 1996), chain shuffling, error-prone PCR and site-directed
mutagenesis (Gram et al., 1992) to further increase affinity and sensitivity. In light
chain shuffling the VH gene of the parental antibody is joined with the random VL gene
of the generated library stock and the newly generated library screened by phage
display. This has been shown to be successful with low affinity antibody fragments
(Low et al., 1996) with 20-fold improvements (Marks et al., 1992). Error prone PCR
involves introducing random nucleotides in the gene sequence and biopanning against
the immunised antigen. It is important not to alter too many nucleotides as it may
influence the binding capabilities of the translated antibody fragment. Site-directed
mutagenesis involves changing specific amino acid sequences within the known
antibody sequence to allow selection of an antibody fragment with improved binding.
27
Tyrosine is an important amino acid in protein recognition and is a common amino acid
in the CDR binding region of the antibody variable heavy chain. It has been shown to
increase antibody binding to protein molecules and is therefore seen as a favourable
amino acid residue. The binding pocket of the antigen should be known prior to sitedirected mutagenesis as key amino acids may be required for antibody/antigen
interaction. These amino acids should not be deleted or substituted as they may hinder
the binding ability of the antibody fragment. A problem arising with mutagenesis
strategies are the library size requirements which may restrict the selection of improved
higher affinity antibodies (Irving et al., 1996). Large library sizes are required to ensure
the retrieval of antibody fragments with improved binding affinity over the parental
clone. By incubating less antigen on the immunotubes the parental antibody fragment
should not be selected or propagated in sequential rounds of biopanning (Smith, 1985).
This approach should ensure the selection of recombinant antibody fragment with
improved binding characteristics towards the antigen.
1.5 Immunoassays for PCa detection
The application of antibodies for the detection of a variety of analytes in medicine,
agriculture and food industries is an area of major significance. It is possible to produce
antibodies against a very broad range of antigens, including conformationally intact or
denatured proteins, short peptides, carbohydrates, drugs, hormones and small chemicals
(Delves et al., 1997). Indeed, immunological detection with antibodies is perhaps the
only technology that has been successfully employed for such a range of analytes.
While the versatility of such assays is widely recognised, advancement in their quality
and usefulness arises from the more recent availability of antibodies with higher
specificity. This achievement can largely be attributed to several key discoveries
involving novel methods of antibody production (Kohler and Milstein, 1975; Krebber et
al., 1997). Today, three main types of antibody are used for immunodiagnostic
applications i.e. polyclonal, monoclonal and recombinant antibodies.
28
1.5.1 Current immunoassays for PSA detection
Advances in immunoassay technology have facilitated the development of sensitive
immunoassays for the measurement of PSA in blood samples to detect the presence of
PCa. PSA is currently the best available marker for PCa detection and is an FDA
approved marker as previously described in Section 1.2.3. The major application of
PSA measurement is for screening men of 45 years and older for the presence of PCa. A
tPSA measurement above a cut-off value of 4.0 ng/mL is generally regarded as positive
and indicates the need for biopsy. PSA testing is also used to monitor the response of
PCa patients to ablative therapy such as radical prostatectomy. In this situation tPSA
should be undetectable, and subsequent detectable levels of tPSA is a sign of disease
recurrence. For this reason, more sensitive PSA assays have been developed to facilitate
earlier detection of recurrence.
Although there are different ways to assemble immunoassays, the most common format
is a solid phase, two-site sandwich immunoenzymatic assay for use on high throughput
immunoassay analysers. The capture antibody is immobilised onto coated paramagnetic
beads, and an enzyme-labelled secondary antibody directed against a distinct antigenic
site allows colorimetric determination of tPSA concentration. The use of paramagnetic
beads as the solid phase allows efficient washing steps, since the beads can be captured
from a solution upon application of a magnetic field, but at they do not themselves
become magnetised they are resuspended upon removal of the magnetic source.
1.5.2 History of PSA immunoassay development for clinical use
Before the distinction of the various forms of tPSA (i.e. free and bound forms) was
established as an important factor in PCa detection, the available antibodies did not
distinguish between such forms, and assays detected tPSA levels only.
The first
immunoassay for the measurement of tPSA was reported in 1980 (Kuriyama et al.,
1980), and the first commercial immunoassays for tPSA used clinically were available
in the late 1980s, namely the Hybritech Tandem-R and the Yang Pros-Check
radioimmunometric assays. These assays were manually performed and had analytical
sensitivities (i.e. the lowest concentration of measured PSA that could be distinguished
29
from zero samples) of the order of 0.5 ng/mL. These ‗first generation‘ assays were
sensitive enough to distinguish elevated PSA from normal levels, but not adequate for
the early detection of disease recurrence following, for example, radical prostatectomy.
During the early to mid 1990s a large body of research both by clinical researchers and
diagnostic immunoassay manufacturers led to the development of numerous
immunoassays with improved limits of detection.
Table 1.2: Commercially available tests commonly used for PSA determinations in
serum
Manufacturer
PSA isoform
Claimed
Claimed
detected
sensitivity
range
(ng/mL)
(ng/mL)
Free PSA
< 0.008
0–30
Total PSA
< 0.008
0–100
Free PSA
< 0.02
0–10
Total PSA
< 0.04
0–50
IMx
Total PSA
< 0.04
0–50
Immunlite
Total PSA
< 0.04
0–150
Products
Free PSA
< 0.02
0–25
Corporation
3rd Generation PSA
< 0.003
0–20
Total PSA
< 0.002
0–100
Free PSA
< 0.01
0–50
Access-
Total PSA
< 0.008
0–150
Hybritech
Free PSA
< 0.005
0–20
ADVIA
PSA-ACT
< 0.03
0–100
Abbott
Assay
Architect
Diagnostics
AxSYM
Diagnostic
Roche
Beckman Coulter
Bayer
Elecsys
Diagnostics
These ‗second generation‘ assays, also termed hypersensitive or ultrasensitive, provided
detection limits in the region of 0.05 ng/mL, i.e. roughly a 10–fold improvement in
analytical sensitivity compared to first-generation assays. At the same time, automated
immunoassay analysers were developed, providing a platform for the processing of
hundreds of samples per day with reduced inherent error compared to earlier manual
30
tests. The first ‗third generation‘ PSA immunoassay was available to the U.S. market in
1997, with the subsequent introduction of numerous other PSA tests in the following
years. Third generation assays offer analytical sensitivities in the region of 0.005
ng/mL, and are generally performed on high throughput analysers allowing the
processing of hundreds or thousands of samples per day if required. Today, there are
numerous PSA assays available in the marketplace (Table 1.2), although it should be
noted that only a limited number have attained FDA approval for clinical use, for
example the Tandem E and Tandem R assays (Formerly Hybritech, now Beckman
Coulter), IMx assay (Abbott), and Immunlite assay (Diagnostic Products Corporation).
Measurement of tPSA using a cut-off level of 4 ng/mL, is associated with a significant
rate of false-positive and false-negative results and is unable to distinguish PCa from
BPH. Therefore, at the same time as assay sensitivity was being improved, more
specific assays were also being developed. The increased assay specificity was based on
the detection of different molecular forms of PSA, including PSA complexed to α1antichymotrypsin (PSA-ACT) and free (uncomplexed) PSA (fPSA), discovered in the
early 1990s (Lilja et al., 1991). In a landmark paper, Catalona et al. demonstrated that
measurement of % fPSA within the so-called diagnostic grey zone of 4–10 ng/mL
significantly improved the specificity of cancer detection compared to tPSA
measurement, such that 20% of unnecessary biopsies could be avoided (Catalona et al.,
1995). By 1998, assays specific for the complexed form of PSA were reported with
sensitivities as low as 0.05 ng/mL (Allard et al., 1998; Wu et al., 1998). A year later an
ELISA capable of detecting the levels of uncomplexed, fPSA in serum was first
reported by Matsumoto and co-workers (Matsumoto et al., 1999). The detection limit
for fPSA established for this assay was 0.008 ng/mL. Today many commercial assays
are available that discriminate between different PSA isoforms, allowing better
discrimination between benign and malignant conditions than by measurement of tPSA
alone. However, studies have shown significant differences in the sensitivities of these
assays, indicating a lack of reliability (Oberpenning et al., 2002). It is for this reason
that current research is still intensely focused on the development of more reliable and
accurate assays for the presence of PCa.
31
One aspect of such research is the development of immunoassays that can demonstrate
higher specificities for the various forms of PSA. Assays able to specifically detect
these isoforms could provide further accuracy in discriminating PCa from benign
disorders (Steuber et al., 2002). In 2000, Nurmikko et al. produced a PSA-specific
monoclonal antibody that did not detect the internally cleaved form of PSA (Nurmikko
et al., 2000).
A year later, the same group reported the development of an
immunofluorometric sandwich assay for the measurement of intact, fPSA, but not
internally cleaved forms (Nurmikko et al., 2001). The detection limit of this assay was
determined to be 0.035 ng/mL. To date as mentioned in Section 1.2.4, limited research
seems to indicate that measurement of the ratio of internally cleaved to normal forms of
PSA may be useful to discriminate benign from malignant prostatic disease (Steuber et
al., 2002). Hence, an immunoassay capable of quantifying these forms could potentially
increase the accuracy of detection of PCa.
1.6 PSA assays for forensic applications
In addition to its use in the diagnosis of PCa, detection of PSA has also become the
method of choice for determining the presence of semen in the absence of sperm and
when sperm count is low, in forensic sexual assault cases. Measurement of PSA in such
samples is associated with many variables not applicable to normal serum samples.
Problems encountered with forensic samples include contamination with other body
fluids or dirt, scarcity of sample, the need for extraction from different types of fabric
leading often to low extraction efficiencies and sample decomposition in burnt cadavers
or following laundry of fabrics. Other problems associated with the detection of PSA
include specificity, false positive/negative results, sensitivity to pH/temperature,
interference from matrix components and PSA stability. In sexual assaults, PSA may be
exposed to harsh environmental conditions for prolonged periods of time that will alter
the protein. This can happen before analysis is carried out and can lead to
conformational changes, reduction of enzymatic activity, and degradation of the
molecule. Furthermore, this can result in non-detection of the molecule by the antibody
used in the test, giving false negative results. The vaginal environment and secretions
are also acidic, which may also contribute to denaturation of PSA. Methods for the
detection of PSA in semen include Ouchterlony double diffusion, crossover
32
electrophoresis, rocket immunoelectrophoresis, radial immunodiffusion and ELISA
(Graves et al., 1985).
PSA immunoassays involving an ELISA format commonly used in tests for clinical
measurement of serum PSA have been studied for their potential use in the forensic
identification of semen (Johnson and Kotowski, 1993; Jimenez-Verdego et al., 1994;
Simich et al., 1999). These high sensitivity assays are relatively time consuming and
expensive for the analysis of small numbers of samples, and unsuitable for forensic
applications because of the aforementioned difficulties associated with forensic
samples. It has been reported that the specificity and sensitivity of ELISA-based PSA
tests becomes compromised for forensic semen samples that have been subjected to
laundry treatment or exposed to contaminants including dishwashing detergents,
shampoo, toothpaste and cleansers (Johnson and Kotowski, 1993).
Also, PSA is
expressed in female tissue at very low levels and, therefore, antibodies which are too
sensitive would not be desired candidates for assay purposes. The possibility of
antibody cross-reactivity with female PSA was investigated by Laux and Custis (2004)
as potential cross-reactivity would lead to false results. This study demonstrated that the
level of PSA in female tissue varies with the tissue type. Current PSA tests used in
detection
of
semen
are
generally
immunochromatographic
membrane
tests
(Hochmeister et al., 1999; Yokota et al., 2001; Maher et al., 2002; Sato et al., 2002;
Khaldi et al., 2004). The sensitivity issues of a commercial immunochromatographic
test were addressed in the study by Laux and Custis (2004) with respect to the presence
of PSA in female body fluids. Levels of PSA reported prior to extraction were 3.7
ng/mL in female urine and 0.53 ng/mL in female serum. Post-extraction, levels were
0.558 ng/mL and 0.08 ng/mL, for urine and serum, respectively. These results indicated
that female PSA levels are negligible in comparison to male PSA (820 000 ng/mL in
semen and 123 000 ng/mL in semen post-extraction).
Validated immunochromatographic membrane tests include the OneStep ABAcard p30
test (Abacus Diagnostics, California, USA), Seratec PSA Semiquant Kit (Seratec,
Gottigen, Germany), BioSign PSA membrane test (Biosign), SMITEST PSA card
(manufacturer) and Onestep PSA test Strip (FF Diagnostics, Cologne, Germany). These
assays are simple in nature, suitable for use on small numbers of samples, give rapid
33
results, and have been reported to be as sensitive and specific as ELISA tests
(Hochmeister et al., 1999). The membrane tests work on the basis that PSA binds to a
mobile monoclonal anti-PSA antibody in the membrane, forming a mobile antigenantibody complex that migrates through the test area due to the capillary effect of the
membrane. A second labelled anti-PSA antibody that is immobilised in the test area
binds the migrating antibody-PSA complex and leads to the formation of a coloured
band in the presence of sufficient PSA (generally 1–4 ng/mL).
Although membrane tests are far less sensitive than the clinically used second or third
generation PSA assays, the nature of forensic PSA detection is such that an assay that is
too sensitive may yield false positive results, because PSA has been detected in many
female tissues and fluids, including breast tissue and milk (Papotti et al., 1989; Yu and
Diamandis, 1995), amniotic fluid (Yu and Diamandis, 1995) and urine (Breul et al.,
1994). Although more suitable than current ELISA tests for use with forensic samples,
membrane tests still suffer from a number of problems related to specificity, false
positives, interference from matrix components and PSA instability (Hochmeister et al.,
1999; Yokota et al., 2001; Khaldi et al., 2004). We envisage that future PSA tests for
use in forensic work will contain antibodies capable of detecting multiple epitopes on
the PSA antigen: for example, a combination of antibodies directed against pH-stable,
temperature stable and detergent stable PSA elements may improve the detection of
PSA in difficult forensic samples. These tests are likely to be delivered in the form of
handheld portable devices, and similar to future PSA tests for clinical use, may
incorporate new assay formats and detection technologies such as those discussed below
in Section 1.7.
1.7 Novel immunoassay formats reported in the literature
Over the past few years many novel-format PSA measurement assays have been
reported in the literature that aim to facilitate more sensitive and specific measurement
of PSA in serum and semen samples (Table 1.3). Research is now focused on more
mechanised, rapid and reliable assays. Biosensor technology has emerged as a useful
tool in immunodetection strategies for a variety of analytes. Biosensors use a
combination of biological receptor compounds (antibody, enzyme, nucleic acid, etc.)
34
and a physical or physico-chemical transducer directing, in most cases, ―real-time‖
observation of a specific biological event e.g. antibody–antigen interaction.
In recent
years, new immunoassay readout strategies have been devised that have potential to
provide greater sensitivity than more conventional approaches. Strategies including
surface plasmon resonance (SPR), immuno-PCR, microcantilevers, quantum dots,
surface-enhanced Raman scattering (SERS), and atomic force microscopy (AFM) have
shown promise in PSA assay formats. Although this section does not document every
assay reported, selected assays of particular interest are included.
Table 1.3: Novel PSA assay formats recently reported in the literature
Detection
Assay format
method
PSA
Reported
isoform
sensitivity
Reference
detected
Surface
Self-assembled
plasmon
monolayer on
resonance
commercial SPR
PSA-ACT
18.1 ng/mL
Cao et al.
(2006);
Total PSA
1 ng/mL
biochip with signal
Huang et al.
(2005);
enhancement using a
0.15 ng/mL
sandwich assay
Besselink et
al. (2004).
format
Surface
Sandwich
plasmon
immunoassay format
fluorescence
on commercial CM5
spectroscopy
(Biacore) chip
Surface-
Sandwich assay using
enhanced
gold nanoparticles
Raman
coated with a Raman
scattering
scatterer
Time-
Fluorescently-labelled
resolved
nanoparticles
Free PSA
80 fM
Yu et al.
(2004).
Free PSA
1 pg/mL
Grubisha et
al. (2003).
Total PSA
7.0 pg/mL
Ye et al.
(2004);
fluorescence
0.4 ng/mL
Ye et al.
(2005);
Free PSA
35
0.83 pg/mL
Huhtinen et
al. (2004);
0.04 pg/mL
Soukka et al.
(2001).
Electrical
Resonance frequency
sensing
shift in a
Total PSA
10 pg/mL
Lee et al.
(2005).
nanomechanical
microcantilever
Free PSA
0.2 ng/mL
Hwang et al.
(2004);
Conductance change
Free PSA
100 fg/mL
in silicon nanowire
sensor chip array
Wu et al.
(2001);
PSA-ACT
100 fg/mL
Zheng et al.
(2005).
Real-time
Sandwich assay with
PCR
DNA label on
Total PSA
4.8 x 105
Lind et al.
molecules
(2005).
100 fg/mL
Drukier et al.
detection antibody
Multi-photon
Sandwich assay with
detection
I-streptavidin on
Total PSA
(2005).
detection antibody
Disposable
Strip format one-step
Free and
assay
lateral flow assay
total PSA
1 ng/mL
FernandezSanchez et al.
(2005).
1.8 Biosensor devices
This approach has many advantages over traditional methods for the detection of minute
amounts of target analyte. It allows label-free, rapid monitoring of biomolecular
reactions in ‗real-time‘, and requires only a small amount of sample. The SPR device
consists of a sensor chip containing flow cells, a light source, a prism and a detector
(Figure 1.10). SPR sensor chips are usually composed of a carboxymethyl dextran
matrix linked to a thin layer (~50 nm) of gold substrate on the sensor surface (Wilson,
2002). SPR measures changes in refractive index and thus in the resonance angle at
which polarised light is reflected from a surface, which is in turn related to a change in
layer thickness or mass (Wilson, 2002). Thus, if polarised light strikes a gold layer at
36
the interface between media (e.g. glass and buffer) of differing refractive indices at a
fixed wavelength above a critical angle, the photons are absorbed into surface plasmons
(electron density waves) resulting in resonance and no reflection of light. If the
refractive index changes at one side of the surface due to immobilisation of, for
example, a ligand to its receptor, the resonance angle is changed and the intensity of the
reflected light is increased (Figure 1.10).
A
B
37
Figure 1.10: Schematic of the typical set-up for a surface plasmon resonance (SPR) biosensor.
SPR detects changes in the refractive index in the immediate vicinity of the surface layer of a
sensor chip. SPR is observed as a shadow in the reflected light from the surface at an angle
dependent on the mass of material at the surface. (A) When antigen is passed over the flow cell
it binds to the receptor resulting in a change in the mass of the surface layer, leading to a shift
in the resonance angle (from I to II in the diagram). This change in signal is proportional to the
amount of ligand bound to the receptor, which enables detection of a ligand and determination
of the ligand–receptor interaction affinities and kinetics. (B) The change in resonance angle can
be monitored non-invasively in ‘real-time’ as a plot of resonance signal (proportional to mass
change) versus time.
Huang and co-workers reported an immunosensor based on the use of a commercial
SPR biosensor (Biacore 2000) for the detection of PSA concentrations as low as 1
ng/mL (Huang et al., 2005). Biacore SPR chips are generally composed of a
carboxymethyl dextran matrix linked to a gold substrate on the sensor surface.
However, self-assembled monolayers (SAMs) have been used as the immobilisation
substrate and have been shown to give better sensitivities than dextran layers (Frederix
et al., 2003). The PSA sandwich assay reported by Huang et al. (2005) used a camel
anti-PSA antibody fragment attached to a mixed SAM of thiols for PSA capture, and a
biotinylated mouse monoclonal anti-PSA antibody and streptavidin-modified gold
nanoparticles for amplification of the detection signal. Cao and co-workers
subsequently reported a SPR-based immunoassay specific for PSA-ACT, in which the
limit of detection was enhanced simply by using a sandwich strategy in which an intact
PSA monoclonal antibody was used as an amplifying agent (Cao et al., 2006). Yu and
co-workers reported the use of a CM5 Biacore chip as a dextran surface on which a
sandwich assay for free PSA was conducted (Yu et al., 2004). The assay combined SPR
and fluorescent labelling, i.e. surface plasmon field-enhanced fluorescence spectroscopy
(SPFS). This technology used surface plasmons to excite the fluorescent label and
produce the signal. The limit of detection observed was 80 fM for a 40 minute contact
time, which is many orders of magnitude lower than detection limits of third generation
PSA assays commercially available today. Related to SPR, surface-enhanced Raman
scattering (SERS) has also been used in the detection of minute amounts of target
molecules. In Raman spectroscopy the light scattered inelastically from a molecule
provides information about the molecule's vibrational quantum states. The Raman effect
38
can be greatly strengthened if the molecules are attached to nm–sized metal structures,
similar to the gold sheets on SPR biosensor chips. SERS has also been used in a
sandwich assay format immunoassay for the detection of PSA (Grubisha et al., 2003).
The assay used tracer nanoparticles created by the physisorption of antibodies and
Raman reporter molecules (RRMs) on gold colloids. It has been speculated that SERS
could be used for the sensitive detection of multiple analytes by immobilising a mixture
of different capture antibodies and using nanoparticles coated with different sets of
tracer antibodies and RRMs (Ni et al., 1999). Such an approach is possible because
Raman bands are sufficiently narrow to reduce the likelihood of spectral overlap when
multiple labels are used.
There has been considerable interest in harnessing the very high sensitivity of ‗realtime‘ PCR for use in diagnostic applications. As early as 1992, Sano and co-workers
described a new technique using PCR for the analysis of protein analytes, which they
called immuno-PCR (Sano et al., 1992). This technique was based on a sandwich assay
format, in which the detection antibody was coupled to a DNA molecule, which was
then amplified after washing steps. The amount of product formed during the
exponential phase of DNA amplification, assessed by gel electrophoresis or PCRELISA, reflected the amount of target protein bound by the capture antibody. However,
the use of real-time PCR for DNA quantification, first demonstrated by Sims and coworkers, led to a significant improvement in sensitivity and accuracy over a wide
concentration range (Sims et al., 2000). Recently, the development of real-time
immuno-PCR systems for the specific detection of PSA was reported (Lind et al.,
2005). The most sensitive assay could detect 4.8 x 105 PSA molecules in a 5 µl sample,
equivalent to a sensitivity of approximately 0.2 pg/mL. ‗Real-time‘ immuno-PCR
assays could be applied to simultaneous multi-analyte detection, since primers binding
specific DNA sequences could be used for each of a number of different detection
antibodies.
Much research into immunoassay development is currently focused on the
miniaturisation of assays for future use in disposable format tests. Some disposable
format PSA tests have recently been reported. A disposable immunosensor based on a
lateral flow format has been successfully designed for the detection of free and total
39
forms of PSA (Fernandez-Sanchez et al., 2005). The lateral flow system was based on a
non-competitive sandwich assay format. The membrane was coated with anti-free-PSA
or anti-total-PSA antibodies with a monoclonal anti-PSA urease conjugate as the
probing molecule. The system consisted of a polymer-coated electrode system to which
each immunostrip was placed in contact. Upon completion of the immunoassay on the
strip, hydrolysis of urea by the urease label generated ammonia and an increase in pH,
which resulted in degradation of the polymer and subsequent determination of the
amount of analyte in the sample, PSA could therefore be quantified. A lower detection
limit of 10 ng/mL of both free and complexed PSA species was observed with this
method.
Considerable interest exists in biosensors that use microelectromechanical system
(MEMS) and nanoelectromechanical system (NEMS) detection technologies (Cherian
et al., 2003). Most nanomechanical sensors use silicon-based cantilevers that bend due
to increased surface stress when the target molecule binds to specific receptors on the
cantilever surface. Lee et al., (2005) reported the development of an immunoassay for
PSA using a piezoresistive nanomechanical microcantelever. Electrical detection, via
surface stress changes, of antigen-antibody binding was achieved through microfabricated self-sensing micro-cantilevers. When binding occurred on a functionalised
gold surface, surface stress was induced resulting in cantilever bending and resistance
change of the piezoresistive layer. The authors reported that the novel electrical
measurement system facilitated label-free detection of PSA with a reported achievable
sensitivity of 10 pg/mL.
The last 30 years have seen a significant improvement in PCa diagnosis, general welfare
and life expectancy of the PCa patient. This has coincided with the discovery of PSA,
the most widely used cancer-associated marker for the disease. Despite its
achievements, some researchers have questioned its value in PCa diagnosis, since
increased serum levels are also associated with non-malignant conditions such as BPH.
Immunoassay development is currently focusing on the detection of the various
isoforms of PSA to increase assay specificity and reduce the rate of false positives and
unnecessary biopsies. While numerous reports of assays for the detection of various
forms have been reported, it is still unclear whether these assays will improve the
40
detection of PCa in the presence of non-malignant disease, and therefore further
research is required. It is likely that the specific detection of PCa will require a panel of
antibodies against multiple markers for PCa including other members of the kallikrein
family of proteases, and even non-protein markers such as genes upregulated in PCa
progression. The advent of proteomics and microarray technology will likely involve
the amalgamation of current antibody-based assays with antibodies against novel
biomarkers monitoring the change in levels of numerous factors thus giving a more
accurate assessment of the disease state and facilitating earlier diagnosis.
1.9 Thesis Outline
The aim of the work presented in this thesis was focused on the generation, selection,
characterisation and application of novel recombinant antibody fragments for the
detection of fPSA. The work involved generating 3 recombinant antibody libraries from
immunised sources and selecting antigen-specific antibody fragments using phagedisplay technology (Chapter 3). Recombinant murine antibody fragments selected from
‗biopanning‘ were reformatted and expressed as scAb for enhanced protein expression
and stability due to problems incurred with poor expression levels of murine scFvs.
Individual clones (murine and avian clones) were ranked using a capture assay approach
on the Biacore system and kinetic analysis performed on the most suitable candidate
clones displaying the greatest percentage stability (Chapter 4). All antibodies were
purified and characterised by ELISA, Biacore analysis, SDS-PAGE and Western
blotting. Epitope mapping using a sandwich assay on the Biacore system revealed
potential clones for immunoassay development.
ELISA and Biacore inhibition-based assays were evaluated. However, low sensitivity
proved to be an issue. A sandwich assay approach was investigated using a biotinylated
murine anti-fPSA scAb and an avian anti-fPSA scFv as a detection antibody. This
approach improved assay sensitivity; female serum samples spiked with fPSA were
assessed to evaluate assay sensitivity in the serum matrix. Serum samples from patients
presenting with PCa were analysed to determine levels of fPSA in the serum sample.
PCa tissue samples were analysed by immunohistochemistry using purified murine and
avian anti-fPSA recombinant antibodies and positive staining was observed (Chapter 5).
41
Chapter 2
Materials & Methods
42
2.1 Materials & Methods
2.1.1 Materials
All reagents were purchased from Sigma-Aldrich, Airton Road, Tallaght, Dublin 24,
Ireland except where otherwise stated below.
Reagent
Supplier
Bacteriological Agar
Cruinn Diagnostics Ltd., Hume Centre,
Yeast
Parkwest Business Park, Nanogor Road,
Tryptone
Dublin 12, Ireland.
DNA ligase
ISIS Ltd., Unit 1 & 2, Ballywaltrim
Helper phage
Business Centre, Boghall Road, Bray, Co.
Restriction enzymes
Wicklow, Ireland.
EZ-Link Biotinylation Kit
Fischer Scientific Ireland, Suite 3, Plaza
212, Blanchardstown Corporate Park 2,
Ballycoolin, Dublin 15, Ireland.
fPSA
Lee Biosolutions Inc., 2924 Mary Avenue,
St. Louis, Missouri, (63144-2730) USA.
Methanol
Lennox, John F. Kennedy Drive, Naas
Road, Dublin 12, Ireland.
PCR primers
Eurofins MWG Operon, 318 Worple
Road, Raynes Park, London, SW20 8QU,
UK.
Perfect Prep gel cleanup kit
Unitech, Unitech House, Magna Drive,
Magna Business Park, City West, Dublin
24, Ireland.
Prestain full range rainbow marker
GE Healthcare Life Sciences, Amersham
Place, Little Chalfont, Buckinghamshire,
HP7 9NA, UK.
Superscript system
Bio-sciences, 3 Charlemont Terrace,
Crofton Road, Dun Laoghaire, Dublin,
Ireland.
Trypsin
AGB Scientific Ltd., Dublin Industrial
43
Estate, Dublin 11, Ireland.
Wizard Plus Mini-prep kit
Medical Supply Co Ltd., Damastown,
Promega cDNA synthesis kit
Mulhuddart, Dublin 15, Ireland.
Promega gel clean up kit
2.1.2 Culture media formulation
Media
Ingredients
2x Tryptone Yeast extract medium (2xTY)
Tryptone
16 g/L
Yeast Extract 10 g/L
Super Broth Medium (SB)
Super Optimal Catabolite Medium (SOC)
NaCl
5 g/L
Tryptone
30 g/L
Yeast
20 g/L
MOPS
10 g/L
Tryptone
20 g/L
Yeast Extract 5 g/L
NaCl
0.5 g/L
KCl
2.5 mM
MgCl2
20 mM
Glucose
20 mM
2.1.3 Buffer compositions
2.1.3.1 Phosphate buffered saline (PBS) (150 mM, pH 7.4)
0.15 M NaCl
2.5 mM KCl
10 mM Na2HPO4
18 mM KH2PO4
All the listed reagents were dissolved in 800 mL of ultra pure H2O, adjusted to pH 7.4
and made up to 1 litre with ultra pure H2O.
44
2.1.3.2 PBS-Tween (PBST)
Tween 20 detergent (Sigma) was added to PBS to give a final concentration of 0.05 %
(v/v).
2.1.3.3 HEPES buffered saline (HBS) (240 mM, pH 7.4)
2.38 g HEPES
8.77 g NaCl
1.27 g EDTA
0.5 mL Tween
HEPES, NaCl and EDTA were dissolved in 800 mL of ultra-pure water. The pH was
adjusted to pH 7.4 by the addition of NaOH. The solution was filtered through a 0.2 µm
filter. The Tween 20 was added to the solution and the solution was degassed.
2.1.3.4 Tris buffered saline (TBS) (150 mM, pH 7.6)
0.05 M Trizma base
0.15 M NaCl
The Trizma base and NaCl were dissolved in 800 mL ultra-pure water. The pH was
adjusted to pH 7.6 using NaOH and made up to 1 litre with ultra pure H2O.
2.1.3.5 Tris-acetic acid-EDTA buffer (TAE) (400 mM, pH 8.3)
A 10X stock solution of TAE buffer, supplied as a proprietary reagent by Sigma, was
diluted to 1 litre with distilled water. All agarose gels for DNA visualisation were
prepared and resolved in 1X TAE buffer.
2.1.3.6 Tris-EDTA buffer (TE) (10 mM, pH 8.0)
0.01 M Trizma hydrochloric acid
0.001 M EDTA
45
2.1.4 Equipment list
Equipment
Supplier
Balances (Chyo JK-180, Mettler PJ300)
Medical Supply Company Ltd,
Damastown, Mulhuddart, Dublin 15,
Ireland.
Biacore 3000TM
Pharmacia Biosensor AB, Uppsala,
CM5 chips
Sweden.
Biometra TGRADIENT PCR machine
AGB Scientific Ltd., Dublin Industrial
Orbital shaker
Estate, Glasnevin, Dublin 9, Ireland.
Millipore Ultrapure H2O Milli-Q academic
Sanyo Orbital Incubator
Bio Rad Gene Pulser Xcell
Alpha Technologies, The Leinster
SDS Bio-Rad Min-Protean® 3 Cell
Technology Centre, Blessington Industrial
Trans-Blot®SD Semi-Dry Transfer cell
Estate, Blessington, Co. Wicklow, Ireland.
Bio-Rad Powerpac Basic
Bio-Rad DNA gel apparatus (Wide-MiniSub® CellGT)
Blood tube rotator SB1
Stuart Scientific, Beacon Road, Stone,
Staffordshire, ST15 0SA, UK.
Gelaire BSB4 laminar unit
Gelman Ltd, 71 Broomhill road, Tallaght
Industrial Estate, Tallaght, Dublin 24,
Ireland.
HermLe Z233MK-2 refrigerated
HermLe Labortechnik GmbH, 25
centrifuge
Siemensstrasse, Wehingen, 78564,
Germany.
Homogeniser (Ultra-Turrax)
Janke & Kunkel IKA-Werk Ultra-Turrax,
Staufen, 79129, Germany.
Camag Linomat 5 (Lateral Flow
Mason Technologies, Greenville Hall, 228
Instrument)
South Circular Road, Dublin 8, Ireland.
New Brunswick Scientific-Excella E24
Incubator Shaker Series (plate shaker)
ND-1000 Spectrophotometer (Nanodrop)
46
Tomy Autoclave SX-700E High Pressure
Steam Sterilizer)
pH meter (Orion 3 star)
Medical Supply Company Ltd,
Damastown, Mulhuddart, Dublin 15,
Ireland.
Plate incubator shaking IKA® MTS 2/4
Lennox, John F. Kennedy Drive, Naas
digital
Road, Dublin 12, Ireland.
Platform shaker STR6
Tecan Saffire2 Plate reader
Tecan Austria GmbH, Untersbergstrasse,
5082 Grödig, Austria.
Water bath (Y6 model)
Grant instruments (Cambridge) Ltd, 29
Station Road, Shepreth, Royston, Herts.,
SG8 6PZ, UK.
2.1.5 Bacterial strains used
All bacterial strains cultured and expressed in different nutrient broth solutions are
listed below.
Bacterial Strain
Product details
Suppliers
E.coli XL-1 Blue cells
Stratagene RecA endA1
Techno-path Ltd.,
gyrA96 thi-1 hsdR17supE44
Holland Road, National
relA1 lac (F proAB lacIq Z
Technology Park,
r
M15 Tn10 (Tet ))
Plassey, Limerick,
Ireland.
E.coli TOP10F' cells
Invitrogen F'[lacIq Tn10(tetR)]
Techno-path Ltd.,
mcrA Δ(mrr-hsdRMS-mcrBC)
Holland Road, National
φ80lacZΔM15 ΔlacX74 deoR
Technology Park,
nupG recA1 araD139 Δ(ara-
Plassey, Limerick,
leu)7697 galU galK rpsL(StrR)
Ireland.
endA1 λE.coli Tuner cells
Novagen – Merck F– ompT
–
hsdSB (rB mB–)
47
gal dcm lacY1
Techno-path Ltd.,
Holland Road, National
Technology Park,
Plassey, Limerick,
Ireland.
2.2 Production of murine recombinant antibody fragments
2.2.1 Preparation of bacterial stocks
All bacterial stocks were streaked on 2xTY agar plates overnight at 37oC for individual
colony isolation. Individual colonies were picked and cultured in 2xTY media overnight
at 37oC with shaking at 200 rpm. The overnight cultures were subcultured into fresh
2xTY medium (Section 2.1.2) and grown until reaching the desired cell density (OD600
nm
~ 0.8). The cultures were centrifuged at 4,000 rpm (3,220 x g) for 10 minutes and
resuspended in sterile glycerol (20% v/v final concentration)-supplemented media. The
cell suspension was transferred aseptically to sterile 1.5 mL centrifugation tubes and
flash frozen using liquid nitrogen. Bacterial strains were stored at -80oC.
2.2.2 Immunisation of Balb/c mice with fPSA and antibody titre determination
The Laboratory Animal Science and Training (LAST) Ireland Animal handling training
course was attended and the associated examination passed. The proposed work was
passed by the ethics committees of the School of Biotechnology and the University and
the licence subsequently granted by the Department of Health and Children. Extreme
care was taken to minimise stress levels to the animals. Balb/c mice aged between 5–8
weeks were used for production of recombinant antibodies to fPSA. These were initially
immunised in the peritoneal cavity with a final concentration of 50 g/mL fPSA in 200
µl PBS, mixed in a 1:1 ratio with Freund‘s complete adjuvant. The mixture was
vigorously shaken to ensure a complete homogenous solution. Freund‘s Complete
Adjuvant (FCA) is an emulsified mineral oil antigen solution used as an
immunopotentiator. Subsequent ‗booster‘ injections (4, 8 and 12 weeks) were
performed with 25 g/mL fPSA mixed in a 1:1 ratio with Freund‘s Incomplete
Adjuvant (FICA). A tail bleed was taken from the mice 7 days after the second booster
administration to confirm the animals were responding to the immunisations and
48
antigen-specific antibodies were being produced. Once the titre reached an appropriate
level (1:100,000 dilution) as determined by ELISA, the mice were boosted 7 days prior
to sacrifice (Krebber et al., 1997).
2.2.3 Extraction and isolation of total RNA from the spleen of an immunised Balb/c
mouse
All work was performed in the Bioresource facility in DCU, Dublin, Ireland. Prior to
use the laminar unit (Gelaire BSB 4) was cleaned with Industrial Methylated Spirits
(IMS) and ‗RNase-free‘ wash solution thus ensuring an uncontaminated environment.
The mouse was sacrificed by cervical dislocation and transfer to the laminar unit. Its
spleen was removed and homogenised in TRI reagent (2 mL) using a pre-baked (180oC,
overnight) sterile homogeniser. The mixture was incubated for 5 min at room
temperature to permit the total dissociation of nucleoprotein complexes. Chloroform
(400 L) was added and the solution shaken vigorously for 15 seconds. The resulting
mixture was stored for 15 min at room temperature and after incubation centrifuged at
12,000 rpm (Eppendorf centrifuge 5810R) for 15 min at 4oC. Three layers were
observed, a lower red phenol/chloroform phase, a protein interphase and a colourless
liquid upper phase containing the RNA. The aqueous upper phase was precipitated
using isopropanol (1 mL) in a fresh ‗RNase-free‘ tube. This mixture was stored for 10
min at room temperature and centrifuged at 12,000 rpm for 8 min at 4oC. RNA forms a
white pellet at the bottom of the tube. The supernatant was removed, the pellet washed
with 75% (v/v) ethanol (2 mL) and centrifuged at 7,500 rpm for 5 min at 4oC
(Eppendorf centrifuge, 5810R). The pellet was allowed to dry and resuspended in
molecular grade water (140 L).
2.2.4 Reverse transcription of total RNA to cDNA
A PCR was performed for 1 hour at 42oC to transcribe total RNA to cDNA. The
synthesis of cDNA was performed using a Promega cDNA synthesis kit containing
oligo (dT) primers. The PCR reaction contained the following components
49
Components
Stock Concentration Final Conc.
Volume (20L)
MgCl2
25 mM
5 mM
4 L
10X Buffer
10 X
1X
2 L
dNTP mix
10 mM
1 mM
2 L
RNase
40 U/L
1 U/L
0.5L
Oligo (dT) Primer
0.5 g/L
0.2 g/rxn
0.4 L
AMV RT*
25 U/L
10 U/rxn
0.4 L
RNA
1.5 g/L
11.25 g/rxn
7.5 L
3.2 L
Molecular Grade H20
* Avian myeloblastosis virus reverse transcriptase
All the components, excluding the enzymes, were placed on ice and slowly thawed. The
enzymes were removed from the -20oC freezer immediately before use to minimise loss
of activity. The components were added to sterile PCR tubes and the PCR programme
initiated.
Temperature (oC)
Time (min)
1
42
15
2
95
5
3
4
pause
Steps
2.2.5 PCR Primers
The primers listed below were obtained from Sigma-Genosys Ltd and are compatible
with the primers set described by Krebber for the pAK vector system (Krebber et al.,
1997). The FLAG tag (gactacaaaGAY) and Bam H1 restriction site (ggatcc) were
incorporated into the designed primers for amplifying the antibody variable domains.
Variable light chain reverse primers
LB1
5gccatggcggactacaaaGAYATCCAGCTGACTCAGCC3
LB2
5gccatggcggactacaaaGAYATTGTTCTCWCCCAGTC3
LB3
5gccatggcggactacaaaGAYATTGTGMTMACTCAGTC3
LB4
5gccatggcggactacaaaGAYATTGTGYTRACACAGTC3
50
LB5
5gccatggcggactacaaaGAYATTGTRATGACMCAGTC3
LB6
5gccatggcggactacaaaGAYATTMAGATRAMCCAGTC3
LB7
5gccatggcggactacaaaGAYATTCAGATGAYDCAGTC3
LB8
5gccatggcggactacaaaGAYATYCAGATGACACAGAC3
LB9
5gccatggcggactacaaaGAYATTGTTCTCAWCCAGTC3
LB10 5gccatggcggactacaaaGAYATTGWGCTSACCCAATC3
LB11 5gccatggcggactacaaaGAYATTSTRATGACCCARTC3
LB12 5gccatggcggactacaaaGAYRTTKTGATGACCCARAC3
LB13 5gccatggcggactacaaaGAYATTGTGATGACBCAGKC3
LB14 5gccatggcggactacaaaGAYATTGTGATAACYCAGGA3
LB15 5gccatggcggactacaaaGAYATTGTGATGACCCAGWT3
LB16 5gccatggcggactacaaaGAYATTGTGATGACACAACC3
LB17 5gccatggcggactacaaaGAYATTTTGCTGACTCAGTC3
LB
5gccatggcggactacaaaGATGCTGTTGTGACTCAGGAATC3
Variable light chain forward primers
LF1
5ggagccgccgccgcc(agaaccaccaccacc)2ACGTTTGATTTCCAGCTTGG3
LF2
5ggagccgccgccgcc(agaaccaccaccacc)2ACGTTTTATTTCCAGCTTGG3
LF4
5ggagccgccgccgcc(agaaccaccaccacc)2ACGTTTTATTTCCAACTTTG3
LF5
5ggagccgccgccgcc(agaaccaccaccacc)2ACGTTTCAGCTCCAGCTTGG3
LF
5ggagccgccgccgcc(agaaccaccaccacc)2ACCTAGGACAGTCAGTTTGG3
Variable heavy chain reverse primers
HB1
5ggcggcggcggctccggtggtggtggatccGAKGTRMAGCTTCAGGAGTTC3
HB2
5ggcggcggcggctccggtggtggtggatccGAGGTBCAGCTBCAGCAGTC3
HB3
5ggcggcggcggctccggtggtggtggatccCAGGTGCAGCTGAAGSASTC3
HB4
5ggcggcggcggctccggtggtggtggatccGAGGTCCARCTGCAACARTC3
HB5
5ggcggcggcggctccggtggtggtggatccCAGGTYCAGCTBCAGCARTC3
HB6
5ggcggcggcggctccggtggtggtggatccCAGGTYCARCTGCAGCAGTC3
HB7
5ggcggcggcggctccggtggtggtggatccCAGGTCCACGTGAAGCAGTC3
HB8
5ggcggcggcggctccggtggtggtggatccGAGGTGAASSTGGTGGAATC3
HB9
5ggcggcggcggctccggtggtggtggatccGAVGTGAWGYTGGTGGAGTC3
51
HB10 5ggcggcggcggctccggtggtggtggatccGAGGTGCAGSKGGTGGAGTC3
HB11 5ggcggcggcggctccggtggtggtggatccGAKGTGCAMCTGGTGGAGTC3
HB12 5ggcggcggcggctccggtggtggtggatccGAGGTGAAGCTGATGGARTC3
HB13 5ggcggcggcggctccggtggtggtggatccGAGGTGCARCTTGTTGAGTC3
HB14 5ggcggcggcggctccggtggtggtggatccGARGTRAAGCTTCTCGAGTC3
HB15 5ggcggcggcggctccggtggtggtggatccGAAGTGAARSTTGAGGAGTC3
HB16 5ggcggcggcggctccggtggtggtggatccCAGGTTACTCTRAAAGWGTSTG3
HB17 5ggcggcggcggctccggtggtggtggatccCAGGTCCAACTVCAGCARCC3
HB18 5ggcggcggcggctccggtggtggtggatccGATGTGAACTTGGAAGTGTC3
HB19 5ggcggcggcggctccggtggtggtggatccGAGGTGAAGGTCATCGAGTC3
Variable heavy chain forward primers
HF1
5ggaattcggcccccgaggcCGAGGAAACGGTGACCGTGGT3
HF2
5ggaattcggcccccgaggcCGAGGAGACTGTGAGAGTGGT3
HF3
5ggaattcggcccccgaggcCGCAGAGACAGTGACCAGAGT3
HF4
5ggaattcggcccccgaggcCGAGGAGACGGTGACTGAGGT3
2.2.6 Components and PCR conditions for antibody variable domain genes
amplification using Krebber-specific primers
The reaction components and volumes for mouse variable chain amplification were as
follows:
Component
20 L volume
cDNA
0.8 L
dNTP mix
0.4 L
5 X Buffers (A, B, D, G)
4.0 L
VL/VH Forward Primer
0.4 L
VH/VL Reverse Primer
0.4 L
H2O
13.2 L
Sigma Red Taq polymerase
0.8 L
52
PCR buffer compositions: buffer A (7.5 mM MgCl2, pH 8.5), buffer B (10 mM MgCl2,
pH 8.5), buffer D (17.5 mM MgCl2, pH 8.5), buffer G (12.5 mM MgCl2, pH 9.0)
(Figure 3.3).
The PCR conditions for the amplification of the mouse antibody variable chain genes by
PCR were as follows:
Steps
Temperature (oC)
Time (seconds)
1
94
60
2
63
55
3
72
60
Repeat Steps 1 3 seven times
4
94
60
5
69
30
6
72
60
Repeat Steps 4 6 twenty three times
7
72
600
8
4
Pause
After successful amplification of both the variable heavy (VH) and variable light (VL)
chain gene fragments, the PCR products were resolved on a 1% (w/v) agarose gel and
purified using the Promega gel clean up kit.
2.2.7 Purification of VH and VL variable gene fragments using the Promega gel clean up
Kit
The VH and VL chain genes were purified from a 1% (w/v) agarose gel by excising the
correct size fragments, 386–440 bp for VH chain, and 375–402 bp for VL chain. The gel
slices were weighed and binding buffer (add three times the volume of the excised gel
fragment) added. The mixture was incubated in a water bath for 10 min at 50oC. The
guanidine isothiocyanate-containing binding buffer allows for sufficient binding of the
DNA to the silica membrane columns. An equal volume of isopropanol to the weight of
the excised fragment was added and the resulting mixture added to a collection column.
53
Isopropanol encourages precipitation of the excised DNA fragment. The column was
centrifuged at 14,000 rpm for 1 min to remove remaining residual buffer and the flowthrough discarded. The column was washed with 750 L wash buffer, centrifuged at
14,000 rpm for 1 min and washed with 250 L wash buffer. The column was
centrifuged at 14,000 rpm for 2 min and the DNA eluted from the column using 30 L
molecular grade H2O into a clean 1.5 mL sterile centrifugation tube. The absorbance of
the purified DNA was measured at 260 nm using the Nanodrop ND-1000
spectrophotometer and the concentration calculated using the Beer-Lambert law.
2.2.8 Splice by Overlap extension PCR (SOE-PCR)
The VH and VL gel purified fragments were joined with a glycine-serine linker (G4S)4
using SOE-PCR to produce an 850 bp DNA fragment. The SOE primers for the 850 bp
amplification were:-
Single chain forward
5ggaattcggcccccgag3
Single chain reverse
5ttactcgcggccagccggccatggcggactaccccg3
Initially the PCR was performed using different VH and VL template ratios with Phusion
polymerase (Finnzymes). The standard 1 x DNA template concentration for each chain
was 10 ng.
The VH and VL ratios for the SOE-PCR amplification were as follows; (i) 10 ng VH: 10
ng VL, (ii) 20 ng VH: 20 ng VL, (iii) 5 ng VH: 5 ng VL, (iv) 5 ng VH: 10 ng VL, (v) 5 ng
VH: 2.5 ng VL, (vi) 5 ng VH: 1.25 ng VL, (vii) 10 ng VH: 5 ng VL, (viii) 2.5 ng VH: 5 ng
VL, (ix) 1.25 ng VH: 5 ng VL. The concentrations of all other components remained
equal to give a final concentration of 20 µL per reaction.
54
20 L volume
Component
Buffer G*
4 L
10 mM dNTP mix
0.4 L
scForward Primer
0.4 L
scReverse Primer
0.4 L
Phusion polymerase
0.2 L
VH chain (10 ng DNA)
2 L
VL chain (10 ng DNA)
2 L
H20
10.6 L
* For composition of buffer G, see page 51.
The reaction was performed using the following PCR conditions:
Steps
Temperature (oC)
1
98
60
2
98
20
3
55
30
4
72
30
Time (seconds)
Repeat Steps 2 4 six times
5
98
20
6
63
30
7
72
30
Repeat Steps 5 7 twenty times
8
72
300
9
4
Pause
A non-specific 1,000 bp band was preferentially amplified over the desired 800 bp scFv
product and, therefore, optimisation was required (Figure 3.4). DMSO and BSA were
added to the PCR mix to decrease the levels of non-specific amplification and improve
concentration of the desired PCR product. A touchdown PCR was incorporated to
facilitate the amplification of the product.
55
L volume
Component
5X Buffer G*
10 L
scForward Primer
1 L
scReverse primer
1 L
dNTP mix
1 L
VH chain
2.66 L
VL chain
2 L
Red Taq Polymerase
2 L
DMSO
-
BSA
-
H20
-
*For composition of Buffer G, see page 51
The PCR amplification was performed using the listed components with varying
concentrations of DMSO, BSA and H2O. The PCR amplified reactions were as follows
(i) supplemented with 5% (v/v) DMSO (ii) supplemented with 50 µg/mL BSA (iii)
supplemented with 5% (v/v) DMSO and 50 µg/mL BSA (iv) no additives (v) positive
control (in-house scFv) and (vi) negative control (no VH and VL DNA template).
The reaction was performed using the following PCR conditions
Steps
Temperature (oC)
1
94
30
2
63
30
3
72
60
Time (seconds)
Repeat Steps 1 3 five times
4
94
60
5
69
30*
6
72
60
Repeat Steps 4 6 thirty times
7
94
60
56
8
50
50
9
72
60
Repeat Steps 7 9 six times
10
72
600
11
4
Pause
*Note: An annealing temperature decrease of 0.5oC after each successive cycle.
The non-specific band was preferentially amplified over the desired product and
therefore the VH and VL DNA template concentrations were increased with a touchdown
PCR amplification. A range of DMSO concentrations (2%, 4%, 6%, 8% and 10% v/v)
were used in the PCR with Red Taq polymerase and Phusion polymerase, respectively.
Component
L volume
5X Buffer G*
10 L
scForward Primer
1 L
scReverse primer
1 L
dNTP mix
1 L
VH chain (20 ng DNA)
4 L
VL chain (20 ng DNA)
4 L
Red Taq/Phusion polymerase
2 L /0.5 L
DMSO
-
H2O
-
*For composition of Buffer G, see page 51
The PCR amplification was performed using the above components with an increased
DNA concentration of VH and VL (20 ng) with varying concentrations of DMSO and
H2O investigating 2 different polymerases in the PCR. The PCR amplified reactions
were as follows (i) supplemented with 2% (v/v) DMSO (ii) supplemented with 4% (v/v)
DMSO (iii) supplemented with 6% (v/v) DMSO (iv) supplemented with 8% (v/v)
DMSO (v) supplemented in 10% (v/v) DMSO (vi) positive control (in-house scFv) and
(vii) negative control (no VH and VL template) for each tested polymerase.
57
The PCR conditions using the Red Taq polymerase were as follows;
Steps
Temperature (oC)
1
94
60
2
69
30*
3
72
60
Time (seconds)
Repeat Steps 1 3 thirty times
4
72
600
5
4
pause
*Note: An annealing temperature decrease of 0.5oC after each successive cycle.
The PCR conditions using the Phusion polymerase were as follows:
Steps
Temperature (oC)
1
98
30
2
98
10
3
69
20*
4
72
30
Time (seconds)
Repeat Steps 2 4 thirty times
5
72
600
6
4
pause
*Note: An annealing temperature decrease of 0.5oC after each successive cycle.
Other polymerases with proof reading capabilities were investigated. A number of
polymerases were added to the PCR reaction to determine their influence on
amplification.
The various polymerases were (i) Red Taq, (ii) Phusion, (iii) Vent (NEB) and (iv) Pfu
(Promega) polymerase. A touchdown PCR was performed using the above polymerases
using the following reaction component compositions:
58
L volume
Component
5X Buffer G*
10 L
scForward Primer
1 L
scReverse primer
1 L
dNTP mix
1 L
VH chain (20 ng DNA)
4 L
VL chain (20 ng DNA)
4 L
Polymerase
0.25 L
DMSO (5 % (v/v))
2.5 L
H20
-
*For composition of Buffer G, see page 51
The PCR amplification was performed using the above components with various
polymerases (Red Taq, Phusion, Vent and Pfu) and H2O. The PCR amplifications were
(i) supplemented with 5% (v/v) DMSO using Red Taq polymerase (ii) supplemented
with 5% (v/v) DMSO using Phusion polymerase (iii) supplemented with 5% (v/v)
DMSO using Vent polymerase in buffer G (iv) supplemented with 5% (v/v) DMSO
using Pfu polymerase in buffer G (v) supplemented in 5% (v/v) DMSO in Vent
polymerase in Vent buffer (vi) supplemented with 5% (v/v) DMSO in Pfu polymerase
in Pfu buffer (vii) positive control (in-house scFv) and (viii) negative control (no VH
and VL template).
The PCR conditions using the Red Taq polymerase were as follows;
Steps
Temperature (oC)
1
94
60
2
69
30*
3
72
60
Time (seconds)
Repeat Steps 1 3 thirty times
4
72
600
5
4
pause
59
*Note: An annealing temperature decrease of 0.5oC after each successive cycle.
The PCR conditions using the Phusion, Vent and Pfu polymerase were as follows;
Steps
Temperature (oC)
1
98
30
2
98
10
3
69
20*
4
72
30
Time (seconds)
Repeat Steps 2 4 thirty times
5
72
600
6
4
pause
*Note: An annealing temperature decrease of 0.5oC after each successive cycle.
The proof reading polymerases did not enhance the amplification of the desired product.
Therefore, the PCR amplification was performed with a number of DMSO
concentrations (3%, 5%, 8% and 12% (v/v)) using Red Taq polymerase and the desired
800 bp band was successfully amplified and large scale amplification performed.
Components
50 L volume
5X buffer G*
10 L
Sc Forward primer
1 L
Sc Reverse primer
1 L
dNTP mix
1 L
VH
4 L
VL
4 L
H2O
24.5 L
5% (v/v) DMSO
2.5 L
Red Taq polymerase
2 L
*For composition of Buffer G, see page 51
The PCR conditions using the Red Taq polymerase were as follows;
60
Steps
Temperature (oC)
1
94
60
2
69
30*
3
72
60
Time (seconds)
Repeat Steps 1 3 thirty times
4
72
600
5
4
pause
The amplified SOE-PCR product (scFv) was separated on a 1% (w/v) agarose gel, DNA
excised and gel purified. The DNA was quantified using the Nanodrop ND-1000
spectrophotometer. The gel-purified DNA was incubated with the restriction enzyme
SfiI (NEB) in parallel with the pAK 100 phagemid vector.
2.2.9 pAK 100 vector purification using Wizard miniprep DNA purification system
A single colony of E. coli XL1-Blue cells harbouring the pAK 100 (all vectors were
donated by Andreas Pluckthun, University of Zurich, Switzerland) was picked from a
streaked agar plate and grown overnight at 37oC at 200 rpm in 2xTY media
supplemented with 25 µg/mL chloramphenicol. The overnight culture was centrifuged
at 4,000 rpm for 10 min at room temperature. The supernatant was discarded and the
pellet thoroughly resuspended in 250 L cell resuspension solution. The suspension was
transferred to a 1.5 mL sterile centrifuge tube and 250 L cell lysis solution added,
inverted 4 times and incubated at room temperature for 5 min. The mixture was
homogenised by inversion and incubated at room temperature for 5 min. Neutralisation
solution containing guanidine hydrochloride, potassium acetate and glacial acetic acid
(350 L) was added and mixed by inverting the tube, as indicated before. This solution
causes the genomic DNA and proteins present in the solution to precipitate. The
bacterial lysate was centrifuged at 14,000 rpm at room temperature for 10 min and the
supernatant-containing plasmid DNA carefully transferred to a spin column and
centrifuged at 14,000 rpm at room temperature for 1 min. The plasmid DNA binds to
the membrane of the column which was washed using 750 L column wash solution
(previously diluted in 100% (v/v) ethanol). The column was re-washed using 250 L
column wash solution and centrifuged at 14,000 rpm at room temperature for 2 min.
61
The DNA membrane spin column was transferred to a new sterile 1.5 mL centrifuge
tube and the plasmid DNA eluted using 100 L molecular grade H2O. The eluted
plasmid DNA was stored at -20oC.
2.2.10 SOE-PCR restriction using SfiI enzyme and ligation into pAK 100 vector for
phage display analysis and ethanol precipitation
The gel-purified scFv fragment and pAK 100 vector for phage selection were cut using
SfiI restriction enzyme. The SfiI restriction enzyme allows for the unidirectional cloning
of scFv fragment into the pAK 100 phage display vector. It recognises 8 bases which
are interrupted by 5 non-recognised nucleotides (5ggccnnnnnggcc3), thus virtually
eliminating internal digestion in antibody sequences. The fragments were incubated at
50oC for 5 hours in a water bath.
The pAK 100 vector and SOE-PCR fragment were digested using SfiI as follows:
40 L volume
Components
SOE-PCR product
1.23 L
NEB 10 X buffer 2
4 L
10 X BSA
4 L
H2O
24.77 L
SfiI
6 L
40 L volume
Components
pAK 100 vector
6 L
NEB 10 X buffer 2
4 L
10 X BSA
4 L
H2O
25 L
SfiI
1 L
Following large scale digestion, the restricted pAK 100 vector and scFv fragment were
gel-purified from a 1% (w/v) agarose gel using the Promega gel clean up kit. The DNA
concentrations were determined on the Nanodrop ND-1000 spectrophotometer and the
62
purified scFv fragment ligated into the purified pAK 100 vector using a ratio of 3:1
(insert to vector) using T4 DNA ligase overnight at 16oC.
The ligation mixture contained the following components;
Components
60 L volume
pAK 100 vector
24 L
scFv fragment
14.62 L
5 X ligase buffer
12 L
H2O
7.28 L
NEB T4 DNA ligase
2 L
After ligation the mixture was ethanol precipitated to remove any salt which would
interfere with the electroporation and cause arcing.
Components
218 L volume
Ligation mixture
60 L
3 M sodium acetate
6 L
100% (v/v) ethanol
150 L
Novagen pellet paint
2 L
The ligation mixture was stored overnight at -20oC. It was centrifuged at 14,000 rpm for
20 minutes at 4oC and the pellet washed using 70% (v/v) ice cold ethanol. The mixture
was centrifuged at 14,000 rpm for 10 min at 4oC and the pellet resuspended in 10 L of
molecular grade H2O. After precipitation the DNA was resuspended in a small volume
of water sterile molecular grade H2O to ensure a high DNA concentration for the
transformation.
2.2.11 Preparation of electrocompetent E. coli cells
The surface, pipettes, shaking incubator, culture flasks, centrifugation bottles and rotor
were thoroughly bleached prior to electrocompetent cell preparation thus ensuring a
clean working environment free of phage contamination. A 10 mL volume of SB
medium supplemented with 30 µg/mL tetracycline was inoculated with a single E. coli
colony picked aseptically from a freshly streaked agar plate and grown overnight at
63
37oC at 200 rpm. Three conical flasks each containing 500 mL of SB with 20% (w/v)
glucose and 1 M MgCl2 were individually inoculated with 2.5 mL aliquots of the
overnight culture. The flasks were shaken at 200 rpm at 37oC until the optical density
(600 nm) reached 0.7. The conical flasks were then incubated on ice and chilled with
the centrifugation bottles at 4oC for 15 minutes. The culture was carefully transferred
into the pre-chilled centrifugation bottles and centrifuged at 4,000 rpm for 20 min at
4oC. The supernatant was removed and the pellets resuspended in 25 mL of ice cold
10% (v/v) glycerol. Two pellets were combined with 10% (v/v) glycerol and
centrifuged at 4,000 rpm for 20 min at 4oC. The pellets were resuspended in 10% (v/v)
glycerol and centrifuged at 4,000 rpm for 20 min at 4oC. The supernatant was removed,
the pellet resuspended in 25 mL 10% (v/v) glycerol, transferred to prechilled 80 mL
polypropylene tubes and centrifuged at 3,500 rpm for 15 min at 4oC. The supernatant
was carefully removed without disturbing the pellet. It was resuspended in 5 mL 10%
(v/v) glycerol, transferred to 1.5 mL centrifugation tubes, flash frozen in liquid nitrogen
and stored at -80oC.
2.2.12 Electro-transformation of XL-1 blue E. coli cells with an scFv-containing
plasmid
An aliquot of electrocompetent XL-1 blue E. coli cells were transformed with the
ethanol precipitated ligation mixture (Section 2.2.10). This was executed using a Gene
Pulser Xcell electroporation system with the controls set at 25 F, 1.25 kV and the
Pulse Controller at 200 . The E. coli cells were thawed on ice and the ligated product
carefully added to the cells, mixed, incubated on ice for 1 min, the mixture was
transferred to an ice-cold 0.2 cm electroporation cuvette. The mixture was gently tapped
to the bottom of the electroporation cuvette. The cuvette was placed in the ShockPod
and the chamber lid closed. A pulse was passed through the suspension in the cuvette
chamber. The cuvette was removed from the chamber and quickly flushed with prewarmed SOC medium (37oC). The electroporated cells were quickly resuspended using
a sterile 1 mL filter tip. The period between applying the pulse and transferring the cells
to the growth media is crucial for recovering E. coli transformants. A delay in transfer,
even by 1 min, causes a 3-fold drop in transformation. The 1 mL suspension was
transferred to a 20 mL sterile centrifugation tube and the cuvette flushed with a further 2
64
mL of SOC media. The transformed cells were shaken for 1 hour at 250 rpm (Sanyo
orbital incubator) at 37oC. Serial dilutions of the transformed library were performed by
diluting the library in SOC media and spreading 100 µL of each dilution on 2xTY plates
supplemented with 25 g/mL chloramphenicol and 1% (w/v) glucose. The 3 mL library
suspension was centrifuged at 4000 rpm for 10 min, resuspended in SOC media and
uniformly spread on 2xTY plates supplemented with 25 g/mL chloramphenicol and
1% (w/v) glucose. A vial containing non-transformed XL-1 blue E. coli cells (negative
control) was plated out on agar plates with 25 g/mL chloramphenicol and 1% (w/v)
glucose. The plates were incubated overnight at 37oC. Individual colonies were counted
on the dilution plates and the library size determined. The transformed colony stock
plates were scraped using a spreader and the suspension centrifuged at 4,000 rpm for 10
min. The cells were resuspended in 20% (v/v) glycerol-supplemented 2xTY media,
transferred to sterile 1.5 mL centrifugation tubes and flash frozen in liquid nitrogen. The
cell library stocks were stored at -80oC.
2.2.13 Rescue of scFv-displaying phages and precipitation of phage-displayed
fragments
To rescue the scFv-displaying phages, a conical flask containing 200 mL 2xTY media
with 2% (w/v) glucose and 25 g/mL chloramphenicol was inoculated with 200 L
from the glycerol library stocks. The flask was shaken at 200 rpm and 37oC until the
mid-exponential phase of growth was reached (OD600 ~ 0.400) and the contents then
centrifuged at 4,000 rpm for 10 min at room temperature. The pellet was resuspended in
fresh 200 mL 2xTY media with chloramphenicol (25 g/mL) and 2 x 1011 pfu/mL
helper phage. The mixture was incubated without shaking for 30 min at 37oC enabling
the initiation of the infection process (F pilus conjugation) of the E. coli cells. The
culture was shaken at 200 rpm at 37oC for 2 hours, 50 g/mL kanamycin (helper phage
resistance gene) added and the culture incubated overnight at 200 rpm at 30oC.
2.2.14 Phage precipitation using PEG/NaCl
The 200 mL culture was transferred to a sterile 250 mL Sorvall and centrifuged at 4,000
rpm for 15 min at 4oC to remove the bacterial pellet. The phage were precipitated by
65
transferring the supernatant to a clean, sterile 250 mL Sorvall centrifuge tube and
adding 8 g of PEG 8,000 (4% w/v) and 6 g of sodium chloride (3% w/v). The
PEG/NaCl was dissolved by shaking the suspension at 180 rpm for 10 min at 37oC. The
250 mL centrifuge tube was placed on ice for 1 hour at 4oC to precipitate and
concentrate the phage-scFv particles. After incubation the centrifugation tube was spun
at 8,000 rpm (7,649 x g) for 25 min at 4oC. The phage scFv pellet was resuspended in 2
mL TE buffer in 2% (w/v) BSA solution. The resuspended phage scFv pellet was
transferred to sterile 1.5 mL centrifuge tubes and centrifuged at 14,000 rpm (19,500 x g)
for 5 min at 4oC. The supernatant containing phage scFv was placed on ice and stored at
4oC until required for selection against the required antigen.
2.2.15 Selection of antigen-binding phage scFv fragments by biopanning using
immobilised antigens
Immunotubes (Nunc, Maxisorp) were coated with 500 L of 50 g/mL fPSA overnight
at 4oC. After blocking with 3% (w/v) BSA in PBS (150 mM, pH 7.4) for 2 hours at
room temperature, 500 L of rescued phage were added and the tubes incubated on a
roller mixer SRT1 (Stuart) for 2 hours at room temperature. The immunotubes were
washed with PBST (x 3) and PBS (x 3). The washing steps remove any non-specific
phage particles that do not bind the immobilised antigen which if not removed could be
propagated in subsequent rounds of selection. PBST removes phage adsorbed due to
non-specific hydrophobic interactions, while PBS removes phage adsorbed by nonspecific ionic interactions. Bound phages were eluted using 500 L of 100 mM glycineHCl, pH 2.2, for 10 min, removed and transferred to a sterile 1.5 mL centrifugation tube
containing 30 L of 2 M Tris-HCl, pH 8.0, for neutralisation purposes. Half of the
eluted phage (265 L) was added to mid-exponential phase (OD600 ~ 0.4) XL-1 Blue
cells and allowed to infect for 30 min at 37oC without shaking for pilus retraction. After
30 min the culture was shaken at 200 rpm for 1 hour at 37oC and centrifuged at 4,000
rpm (3,824 x g) for 10 minutes at room temperature. The pellet was resuspended in 600
L fresh 2xTY media and spread on 2xTY agar plates supplemented with 25 µg/mL
chloramphenicol and 1% (w/v) glucose and incubated overnight at 37oC at 200 rpm.
Input and output titres for each round of selection were performed by infecting E. coli
cells and performing serial dilutions of the phage. The infected cells were incubated for
66
15 min at 37oC and 100 µL of each dilution spread on 2xTY agar plates containing 25
g/mL chloramphenicol and 1% (w/v) glucose. The serial dilution plates were
incubated overnight at 37oC.
The input and output titres were calculated by counting the number of cells present on
the dilution plates. The stock plates containing the eluted phage were scraped using
2xTY media for the next round of selection. The infected cells were grown in 100 mL
of 2xTY media with 2% (v/v) glucose and 25 g/mL chloramphenicol until the midexponential phase of growth was attained (OD600 ~ 0.4). The enriched cells were
centrifuged and resuspended in glucose-free 2xTY media supplemented with helper
phage. The helper phage were allowed infect the E. coli cells for 30 min at 37oC and the
culture vigorously shaken for 2 hours at 37oC. After 2 hours, kanamycin was added and
the culture was shaken at 200 rpm overnight at 30oC. In the second round of selection
the antigen coating concentration on the surface of the immunotubes for antigen binding
phage remained the same (500 µL of 50 µg/mL fPSA) but more stringent washes (PBST
(x 5) and PBS (x 5)) were utilised to reduce non-specific binding of phage to the
immobilised antigen. The phage were eluted in the same manner as described in the
previous round of selection. Input and output phage plates were incubated at 37oC and
XL-1 Blue cells infected with half of the eluted phage, incubated for 30 min at 37oC
without shaking initially, then with shaking for 1 hour at 37oC and the cells were
subsequently uniformly spread on 2xTY agar plates. The third round of selection
involved using 100 mL 2xTY media containing 2% (v/v) glucose and 25 g/mL
chloramphenicol with less antigen (500 L of 10 g/mL fPSA) coated on the surface of
the immunotubes and an increased number of washes (PBST (x 10) and PBS (x 10) to
enhance the selection process and ensuring enrichment of positive clones. The fourth
round of selection involved rescuing the phage-scFv fragments in 100 mL of 2xTY
media with 2% (v/v) glucose and 25 g/mL chloramphenicol and biopanning against
500 L of 2 g/mL fPSA immobilised on the surface of the immunotubes with a further
increase in the number of washing steps (PBST (x 10–15) and PBS (x 10–15)). After
washing, an antigen challenge was performed which involved incubating fPSA (500 µL
of 10 µg/mL) in the immunotubes for 10 min to remove any weakly binding phage-scFv
fragments. Phage-scFv particles bound to the fPSA-coated immunotube were eluted and
the phage infected into mid-exponential phase E. coli cells. During the fifth round of
67
biopanning, the immunotube was coated with 500 µL of 5 g/mL fPSA and washed
(PBST (x 13) and PBS (x 13)).
2.2.16 Polyclonal and monoclonal phage ELISA and scFv check.
An ELISA plate was incubated with 100 L of 2 g/mL fPSA overnight at 4oC. The
fPSA was removed from the wells and a blocking solution of 3% (w/v) BSA in PBS
(150 mM, pH 7.4) added to each well, plates were incubated for 1 hour at 37oC. The
cells containing the rescued phage from the original library and each round of selection
were grown in 10 mL 2xTY media until a culture OD600 ~ 0.4 was reached, and then
centrifuged. The cells were resuspended in 2xTY media supplemented with helper
phage, incubated for 30 min at 37oC without shaking, shaken at 200 rpm for 2 hour at
37oC, kanamycin added and cultures incubated overnight at 30oC at 200 rpm to ensure
protein expression. The cultures were centrifuged at 4,000 rpm (3,824 x g) for 10 min at
room temperature and the phage supernatant diluted 10-fold in 1% (w/v) BSA PBST.
One hundred L of diluted phage supernatant were added in duplicate to the plate and
the plate was incubated for 1 hour at 37oC. After incubation the wells were washed with
PBST (x 3) and PBS (x 3). After washing, 100 L of 1/3,000 dilution of a HRPconjugated anti-M13 antibody (Pharmacia) in 1% (w/v) BSA PBST were added and the
plate incubated for 1 hour at 37oC. Binding of HRP-conjugated anti-M13 antibody was
detected using TMB as substrate for HRP. This solution was left to develop for 20 min
and the reaction quenched using 50 L of 10% (v/v) HCl. The absorbance was then read
at 450 nm.
Colonies (20) were randomly picked from the fifth round of biopanning and a colony
pick PCR performed to ensure the presence of the scFv insert.
The PCR amplification was performed as follows:
68
Volume (154 L)
Components
Buffer G
88 L
sc forward primer
8.8 L
sc reverse primer
8.8 L
DMSO
22 L
dNTPs
8.8 L
Red Taq polymerase
17.6 L
Individual colonies were picked using sterile tips and mixed with the PCR components
(above) and a PCR profile executed as follows:
Steps
Temperature (oC)
1
94
60
2
69
30
3
72
60
Time (seconds)
Repeat Steps 13 thirty times
4
72
600
5
4
Pause
The amplified scFv fragments were resolved on a 1% (w/v) agarose gel. The polyclonal
phage ELISA showed the high absorbance signal from round 5 of biopanning.
Therefore, a monoclonal phage ELISA was performed on individual colonies selected
from rounds 4 and 5 of biopanning. Individual colonies were grown in 200 L 2xTY
media supplemented with 2% (v/v) glucose and 25 g/mL chloramphenicol in sterile 96
well plates until the mid-exponential growth phase. The plates were centrifuged at 4,000
rpm for 10 min at room temperature and the pellets resuspended in fresh 2xTY media
(180 L) containing 25 g/mL chloramphenicol and helper phage. After 30 minutes
incubation at 37oC the cultures were shaken for 2 hours at 200 rpm at 37oC. After the 2
hour incubation period 2xTY media (20 L) supplemented with 500 g/mL kanamycin
was added to each well and incubated overnight at 30oC.
69
An ELISA plate, coated overnight with 100 L of 2 g/mL fPSA, was blocked with 200
L of 3% (w/v) BSA in PBS (150 mM, pH 7.4) for 1 hour at 37oC. The overnight
phage-scFv expressed cultures were centrifuged at 4,000 rpm for 10 min at room
temperature. One hundred µL of the supernatant was incubated with 100 µL of 1%
(w/v) BSA in PBST for 1 hour at 37oC. The pre-blocked phage (100 L) were added to
single wells for 1 hour at 37oC. Following washing with PBST (x 3) and PBS (x 3), 100
L of HRP-conjugated anti-M13 antibody (1:3,000 dilution) were added to each well
and the plate incubated for 1 hour at 37oC. The wells were washed using PBST (x 3)
and PBS (x 3). The substrate, TMB, was added, incubated for 20 min and quenched
using 10% (v/v) HCl. The absorbance was read at 450 nm.
2.2.17 Inhibition phage ELISA
Three individual clones were grown in 20 mL 2xTY media containing 2% (v/v) glucose
and 25 g/mL chloramphenicol until the mid-exponential phase of growth was reached.
The cultures were centrifuged at 4,000 rpm for 10 min at room temperature and the
pellet resuspended in fresh 2xTY media (20 mL) supplemented with 40 L commercial
helper phage (NEB) and 25 g/mL chloramphenicol. The mixture was placed at 37oC
for 30 minutes without shaking and then shaken at 200 rpm for 2 hours at 37oC. The
antibiotic kanamycin (20 µl of 50 µg/mL) was added, and the cultures shaken overnight
at 30oC. An ELISA plate coated overnight with 100 L of 2 g/mL fPSA was blocked
using 3% (w/v) BSA in PBS for 1 hour at 37oC. The cultures containing individual
phage scFv clones were centrifuged at 4,000 rpm for 10 min at room temperature. The
supernatant was incubated with various fPSA concentrations in 1% (w/v) BSA in PBS
(final concentration 5 g/mL, 1 g/mL and 200 ng/mL fPSA) for 30 min at room
temperature in 1.5 mL centrifugation tubes. Varying fPSA concentrations were
incubated with the phage-scFv supernatant to determine if the selected clones were
binding fPSA in solution. The supernatant for each clone containing no fPSA was
diluted to ensure the same concentration for each selected clone. The phage
supernatants (100 L) were added in triplicate to the ELISA plate and incubated for 1
hour at 37oC. Any phage binding to the fPSA in solution was removed by the washing
steps (PBST (x3) and PBS (x3)). An anti-M13 bacteriophage HRP-labelled antibody
(1/3,000 dilution) was added to detect any phage-scFv fragments binding positively to
70
the adsorbed fPSA on the surface of the ELISA plate. The substrate, TMB, was added
and the reaction quenched after 20 min by adding 50 L of 10% (v/v) HCl. The
absorbance was read at 450 nm.
2.2.18 ScAb expression vectors (pMoPAC 16 & 53), plasmid preparation and gel
purification
The pMoPAC series of vector DNA were kindly donated by Dr. Andrew Hayhurst
(Department of Virology and Immunology, Southwest Foundation for Biomedical
Research, 7620 NW Loop 410, San Antonio, TX 78227, USA). The DNA was
transformed into E. coli XL-1 Blue cells using the Gene Pulser using the same
procedure as previously described (Section 2.2.12). The transformed cells were rescued,
spread uniformly on 2xTY agar plates, containing 25 g/mL carbenicillin and 2% (w/v)
glucose, and incubated overnight at 37oC. A single colony containing the plasmid DNA
was picked, used to inoculate 2xTY media containing 25 g/mL carbenicillin, cultures
were grown overnight at 37oC. The plasmid preparation was performed using a Wizard
miniprep kit as described in Section 2.2.9. Vector DNA was stored at -20oC until
required.
2.2.19 Vector and SOE-PCR product digestion
The plasmid DNA and the scFv fragment were cut by incubation with the restriction
enzyme SfiI for 5 hours at 50oC in a water bath. For soluble expression the scFv was
subcloned into the pMoPAC vector from the pAK 100 phagemid vector. Plasmid
purification was carried out on an overnight culture of the pan 5 rescue glycerol cell
stock. The E. coli cells containing the scFv fragments from the fifth round of selection
were chosen due to the large number of positive fragments represented in the polyclonal
phage pool as determined by ELISA (Figure 3.9) The plasmid DNA was subsequently
amplified by PCR using the SOE primers to generate the scFv fragment.
The murine anti-fPSA scFv was amplified from the plasmid DNA using the following
components and PCR conditions:
71
Volume (500 L)
Components
Steps
Plasmid DNA
20 L
Buffer G
50 L
sc forward primer
10 L
sc reverse primer
10 L
dNTPs
10 L
H2O
365 L
3% (v/v) DMSO
15 L
Red Taq polymerase
20 L
Temperature (oC)
Time (seconds)
1
94
60
2
69
30*
3
72
60
Repeat Steps 13 thirty times
4
72
600
5
4
Pause
* 0.5oC decrease in annealing temperature after each consecutive completed cycle.
The amplified scFv was gel-purified from a 1% (w/v) agarose gel and DNA quantified
using the Nanodrop ND-1000 spectrophotometer. The recombinant pMoPAC vector
DNA and the gel-purified scFv were digested using the restriction enzyme, SfiI, for 5
hours at 50oC in a water bath. After digestion the fragments were resolved on a 1%
(w/v) agarose gel and gel purified. The DNA was quantified using the Nanodrop ND1000 spectrophotometer and the scFv subcloned into the pMoPAC vectors.
72
Volume (100 L)
Components
pMoPAC 16 & 53 Vector
50 L
NEB 10 X buffer 2
10 L
10 X BSA
10 L
H2O
20 L
SfiI restriction enzyme
10 L
Volume (50 L)
Components
Mouse scFv insert
16.88 L
NEB 10 X buffer 2
5 L
10 X BSA
5 L
H2O
7.12 L
SfiI restriction enzyme
16 L
2.2.20 Sub-cloning the restricted scFv fragment into the restricted vector fragments for
scAb expression
The gel-purified SfiI restricted scFv insert fragment was subcloned into scAb soluble
expression vectors overnight.
Components
Volume (30 L)
pMoPAC 16 & 53 vector
4.66 L & 4.57 L
scFv fragment
12 L
NEB 5 X ligase buffer
6 L
H2O
2.34 L & 2.43 L
T4 ligase enzyme
5 L
The ligated product was ethanol-precipitated and transformed into Tuner cells (Novagen
Section 2.1.5) which were previously transformed with the pRARE plasmid (supplied
by Dr. Paul Clarke, DCU, Ireland). The transformed cells were rescued using SOC
media (3 mL) with shaking for 1 hour at 200 rpm (37oC). Serial dilutions were
uniformly spread on 2xTY agar plates supplemented with 25 g/mL carbenicillin
73
(pMoPAC vector resistance) and stock plates incubated overnight at 37oC. The stock
plates were scraped to prepare glycerol stocks which were stored in the -80oC freezer
until required.
2.2.21 Soluble expression of scAb fragments in pMoPAC 16 and pMoPAC 53 vectors
Individual colonies were picked and grown in 2xTY media supplemented with 25
g/mL carbenicillin overnight at 180 rpm at 37oC, in sterile 96 well plates. These
colonies were used as back-up stocks for each positive clone selected. Twenty L from
the overnight cultures were inoculated into fresh 2xTY media (180 L) containing 0.5
% (v/v) glycerol and 0.05 % (v/v) glucose (final concentration) (1 x 505 – allows autoinduction of cells without changing growth media) (Studier et al., 2005), 1mM MgSO4
and 25 g/mL carbenicillin. The sterile 96 well plates were shaken at 37oC at 180 rpm
until a cell density of ~0.600 (cells appear turbid) was achieved. A final concentration
of 1 mM IPTG was added to each individual well and plates were shaken overnight at
180 rpm at 30oC. An ELISA plate coated overnight with 100 L of 1 g/mL fPSA was
blocked with 3% (w/v) BSA in PBS for 1 hour at 37oC. The expressed antibody culture
lysate was extracted by placing the plates at -80oC for 10 min until frozen and then
thawing the plates at 37oC. This freeze-thaw method of extraction bursts the E. coli cells
releasing periplasmic proteins. This procedure was repeated 3 times in total and the
plates centrifuged at 4000 rpm for 10 minutes at room temperature. One hundred L of
scAb was added to the ELISA plate and incubated for 1 hour at 37 oC. The plates were
washed using PBST (x 3) and PBS (x 3) to remove non-specific proteins. The scAb was
detected using 100 L of anti-human kappa C light chain antibody labelled with
alkaline phosphatase (1/2000 dilution). The substrate, 4pNPP, was added, the plate
incubated for 30 minutes at 37oC and the absorbance read at 405 nm.
74
2.3 Generation of chicken-derived anti-fPSA recombinant antibody library
2.3.1 Removal of spleens from chickens immunised with fPSA, RNA extraction and
cDNA synthesis
Chickens (Leghorn females) were immunised between the ages of 5–8 weeks. When the
titre was deemed sufficient (after 3rd booster administration with over 1:100,000
response determined by ELISA) for recombinant antibody library generation the spleens
(2) were removed, RNA extracted, as described in Section 2.2.3, and cDNA synthesised
by reverse transcription, as outlined in Section 2.2.4. The cDNA acted as the template
for the generation of both chicken anti-fPSA scFv and Fab antibodies by PCR.
Preparation of the chicken anti-fPSA scFv library involved amplifying the VH and VL
chains separately and linking them with a glycine-serine linker, as described in Section
2.2.8. The generation of the chicken anti-fPSA Fab library was more complicated and
involved a number of essential steps:
(i)
amplifying the VH and VL chain genes separately,
(ii)
performing the first PCR-overlap of the VH with the constant heavy chain (CH)
and VL with the constant light chain (CL),
(iii)
final PCR-overlap of the VH/ CH with VL/ CL to generate the Fab.
2.3.2 PCR Primers
VH Primers
CHybVH (forward)
5-GCT GCC CAA GCC ATG GCC GCC GTG ACG TTG GAC GAG TCC 3
CHybIg-B (reverse)
5-CGA TGG GCC CTT GGT GGA GGC GGA GGA GAC GAT GAC TTC GGT
CCC 3
Vλ Primers
CSCVK (forward)
5-GTG GCC CAG GCG GCC CTG ACT CAG CCG TCC TCG GTG TC 3
CHybL-B (reverse)
75
5-AGA TGG TGC AGC CAC AGT TCG TAG GAC GGT CAG GGT TGT CCC
GGC 3
CH Primers
HIgGCH1-F (forward)
5-GCC TCC ACC AAG GGC CCA TCG GTC 3
dpseq (reverse)
5-AGA AGC GTA GTC CGG AAC GTC 3
CL Primers
HKC-F (forward)
5-CGA ACT GTG GCT GCA CCA TCT GTC 3
Lead-B (reverse)
5-GGC CAT GGC TGG TTG GGC AGC 3
VH /CH Overlap Primers
LeadVH (forward)
5-CGA ACT GTG GCT GCA CCA TCT GTC 3
Dpseq (reverse)
5-AGA AGC GTA GTC CGG AAC GTC 3
Vl /Cl Overlap Primers
CSC-F (forward)
5-GAG GAG GAG GAG GAG GAG GTG GCC CAG GCG GCC CTG ACT CAG 3
Lead-B (reverse)
5-GGC CAT GGC TGG TTG GGC AGC 3
Primers for assembly of chimeric heavy chain with chimeric light chain sequence
CSC-F (forward)
5-GAG GAG GAG GAG GAG GAG GTG GCC CAG GCG GCC CTG ACT CAG 3
dp-Ex (reverse)
5-GAG GAG GAG GAG GAG GAG AGA AGC GTA GTC CGG AAC GTC 3
76
The primers listed above were obtained from Eurofins MWG (Anzinggerstr. 7a, 85560,
Ebersberg, Germany) and are compatible with the primers set described by Barbas for
the pComb vector system (Barbas et al., 2000).
2.3.3 Amplification of VH and VL chain genes for Fab library generation
Each PCR amplified chain (VH or VL) was resolved on an agarose gel, gel-purified and
the purified DNA quantified using the Nanodrop ND-1000 spectrophotometer. The
variable region genes were amplified using different concentrations of MgCl2 (1.5, 2.0,
2.5 and 3.0 mM), respectively. MgCl2 promotes activity of the DNA polymerase and is
a crucial component for successful amplification. The VH and Vλ chain genes were
amplified by PCR using the following components and conditions:
Component
L volume
cDNA
1 L
dNTP
1 L
5 X Buffer (Promega)
5 L
Forward Primer
1 L
Reverse Primer
1 L
H2O
-
Taq polymerase
0.25 L
MgCl2
-
Steps
Temperature (oC)
1
94
300
2
94
15
3
56
15
4
72
90
Time (seconds)
Repeat Steps 2 4 thirty times
5
72
600
77
The constant heavy and light chains were amplified from the pComb3XTT vector by a
co-worker (Dr. William Finlay) and used in the first SOE-PCR by joining the VH chain
with the CH and VL chain with the CL chain genes.
Component
20 L volume
Variable region
1 L
Constant region
1 µL
dNTP
1 L
5 X Buffer (Promega)
5 L
Forward Primer
1 L
Reverse Primer
1 L
H2O
6.75 µL
Taq polymerase
0.25 L
MgCl2
3 µL
Steps
Temperature (oC)
1
94
300
2
94
15
3
56
15
4
72
120
Time (seconds)
Repeat Steps 2 4 fifteen times
5
72
600
The heavy and light chain constructs were gel-purified, ethanol-precipitated and
quantified using the Nanodrop ND-1000 spectrophotometer. The final SOE-PCR
involved joining the 2 gel-purified domains to generate the Fab construct. The final
overlap was performed using a Platinum Taq high fidelity polymerase (Bio Sciences
Ltd., 3 Charlemont Terrace, Crofton Road, Dun Laoghaire, Dublin, Ireland) to eliminate
the introduction of mutations in the construct which may inhibit binding to the antigen.
78
Component
50 L volume
Heavy chain region
1 L
Light chain region
1 µL
dNTP
1 L
10 X Buffer
5 L
Sense Primer
0.5 L
Reverse Primer
0.5 L
H2O
38.8 µL
Platinum Taq polymerase
0.2 L
MgSO4
2 µL
Steps
Temperature (oC)
1
94
300
2
94
15
3
56
15
4
72
180
Time (seconds)
Repeat Steps 2 4 Twenty five times
5
72
600
After large scale amplification, the construct was resolved on a 2% (w/v) agarose gel,
purified and quantified using the Nanodrop ND-1000 spectrophotometer. The next stage
involved SfiI digestion of both the gel-purified Fab construct and pComb3XSS
phagemid vector at 37oC for 5 hours.
Component
200 L volume
Fab construct
60 µL
10 x buffer 2
20 µL
10 x BSA
2 µL
SfiI
8 µL
H20
110 µL
79
Component
200 L volume
pComb3XSS
60 µL
10 x buffer 2
20 µL
10 x BSA
2 µL
SfiI
8 µL
H20
110 µL
After gel purification, the digested overlap PCR product was ligated into the
pComb3XSS vector (supplied by Barbas). The ligation was performed overnight at
room temperature and the plasmid DNA ethanol precipitated at -20oC. The ethanol
precipitate was centrifuged at 4oC and the pellet resuspended in molecular grade H2O.
Component
400 L volume
pComb3XSS vector
120 µL
Fab construct
90 µL
5 x ligase buffer
80 µL
Ligase enzyme
10 µL
H2O
100 µL
2.3.4 Transformation into XL-1 Blue E. coli cells
The ligation mixture was transformed into electrocompetent E. coli XL-1 Blue cells and
the library size determined by counting the number of colonies from plate dilutions.
Both chicken anti-fPSA scFv and Fab antibodies were selected by biopanning the phage
against fPSA adsorbed onto the wells of ELISA plates. In the first round of biopanning
8 wells of an ELISA plate were coated with 100 L of fPSA (10 g/mL), blocked with
3% (w/v) BSA in PBS (150 mM, pH 7.4), and the eluted phage, after PEG precipitation,
added to each well and allowed to bind for 1 hour at 37oC. Each well that contained
phage was washed with PBST (x 3) and PBS (x 3). After washing 100 L of 10 mg/mL
trypsin in PBS (150 mM, pH 7.4), was added to each well and the plate incubated for 30
min at 37oC. The pComb 3XSS vector has a trypsin-sensitive site that is cut following
the addition of trypsin. Phage were aspirated from the wells by pipetting the content of
the wells vigorously i.e. removing the liquid in each and releasing it again into the well.
80
Half of the eluted phage (400 L) were added to mid-exponential phase E. coli XL-1
Blue cells. The culture was placed at 37oC for 15 min and then shaken for 1 hour at 200
rpm and 37oC. Input and output phage titres were determined, as previously described in
the mouse anti-fPSA scFv biopanning protocol. In round two of selection, 2 ELISA
plate wells were coated with 100 L of fPSA (10 g/mL) and in round three of
selection, 1 ELISA plate well was coated with 100 L of fPSA (10 g/mL). The
numbers of washing steps for each round of selection were PBST (x 5–10), PBS (x 5–
10 second round) and PBST (x 10–15) and PBS (x 10–15 third round), respectively.
Gradually decreasing the concentration of antigen and increasing the washing
stringency encourages the selection of positive antibody recombinant fragments from
the generated library. Phage were eluted as before using 100 L of 10 mg/mL trypsin
per well and used to infect E. coli XL-1 Blue cells. After each round of selection the
infected E. coli cells were spread on 2xTY agar plates containing 100 g/mL
carbenicillin and 2% (v/v) glucose. Individual colonies (5) from each round of
biopanning were selected and a scFv check performed. This involved picking a colony
and performing a scFv-specific PCR. The components and PCR conditions were as
follows:
Volume (20 L)
Components
Steps
Forward primer
0.2 L
Reverse primer
0.2 L
dNTPs
0.4 L
Buffer G
2 L
H2O
16.4 L
Red Taq polymerase
0.8 L
Temperature (oC)
Time (seconds)
1
94
60
2
69
30
3
72
60
Repeat Steps 13 thirty times
4
72
600
5
4
Pause
81
A polyclonal phage ELISA was performed, as described previously (Section 2.2.16),
and colonies from round 3 of biopanning analysed for soluble expression of chicken
anti-fPSA scFv fragments. Individual clones were selected from 2xTY agar plates,
supplemented with 100 g/mL carbenicillin, and functionally expressed. The positive
clones binding to fPSA on the surface of the ELISA plate were subjected to further
examination. A DNA fingerprinting assay was performed on 5 anti-fPSA Fab clones
selected from the original library and 5 clones from the final round of selection. The
clones were cut using a high frequency cutting enzyme, AluI to check for diversity. The
DNA from the clones was resolved on a 3% (w/v) agarose gel. The eluted phage from
both the chicken anti-fPSA scFv and Fab antibodies were used to infect E. coli
TOP10F cells for soluble expression of the antibodies. Individual clones were picked,
grown overnight at 37oC, sub-cultured into fresh 2xTY media supplemented with 1 x
505 (50 % (v/v) glycerol and 5 % (v/v) glucose solution) and 1 mM MgSO4 and grown
until turbid (OD600 ~ 0.6). Media supplemented with IPTG (1mM) was added to each
clone and the plate shaken overnight at 30oC. Antibodies expressed in the periplasmic
space were extracted from the E. coli cells using the freeze-thaw method and individual
clones analysed by ELISA. The positive clones were stored in 2xTY media
supplemented with 20% (v/v) glycerol at -80oC. High throughput screening of selected
positive anti-fPSA clones were evaluated on the Biacore A100 by Dr. Paul Leonard and
the data analysed.
2.4 Murine anti-fPSA scAb analysis using the Biacore 3000 instrument
Analysis of murine anti-fPSA scAb was performed on a Biacore 3000TM instrument
using a CM5 dextran chip. The running buffer was HBS buffer, pH 7.4, containing 10
mM HEPES, 150 mM NaCl, 3.4 M EDTA, and 0.05% (v/v) Tween 20. The running
buffer was passed through a 0.22 µm nylon filter and degassed using a Millipore
filtration unit directly before use.
2.4.1 Mouse monoclonal anti-human kappa light chain purification
The mouse monoclonal anti-human kappa light chain clone KP-53, supplied in ascites
fluid from Sigma, was purified by Protein G affinity chromatography. The Protein G82
Sepharose 4B fast flow slurry resin (500 µL) was equilibrated using 20 mL of PBS (300
mM, pH 7.4). An aliquot (200 L) of anti-human kappa light chain monoclonal
antibody, diluted in 5 mL of PBS (300 mM, pH 7.4) was added to the resin. The
solution was passed through the resin and re-circulated 5 times in total. The resin was
washed using 20 mL of PBS (300 mM, pH 7.4), to remove the ascites fluid. The
captured monoclonal antibody was eluted in 0.1 M glycine-HCl solution, pH 2.4.
Collected fractions were neutralised with 1 M Tris-HCl solution, pH 7.5. The fractions
were pooled and passed through a PD-10 desalting column (removal of low-molecular
weight compounds) to remove any NaCl residue that may affect downstream processing
(for successful immobilisation the ligand must be in a pure form). The PD-10 column
was equilibrated with 20 mL H2O and the eluted protein G-purified fraction applied to
the resin. The anti-human kappa light chain mAb passes quickly through the resin by
size exclusion; 500 L fractions were collected. An Immunodot blot was performed
using an AP-labelled anti-mouse antibody to determine the fractions containing the
purified antibody. The fractions were pooled and centrifuged in a Millipore Amicon
ultrafiltration device (Vivaspin 6 – 10,000 molecular weight cut-off) to ensure a high
concentration for preconcentration studies and immobilisation.
2.4.2 Biacore maintenance
The Biacore instrument was cleaned using a ‗super desorb‘ program at 25oC. This
method involves cleaning the unit thoroughly prior to starting an assay. It is imperative
that the instrument is properly sanitized to remove any residual protein, thus, ensuring
high-quality data and complete confidence in results generated. A maintenance chip was
docked and primed 5 times with desorb solution 1 (0.5 % v/v sodium dodecyl sulphate
(SDS)), once with water, 5 times with desorb solution 2 (50 mM glycine, pH 9.5), and,
finally, 5 times with water.
2.4.3 Preconditioning of the CM5 dextran sensor chip
After docking the CM5 chip, the dextran layered surface was cleaned and hydrated
using 2 consecutive injections each of 100 mM HCl, 50 mM NaOH and 0.5% (v/v)
83
SDS. These solutions were passed over each of the 4 flow cells for 10 seconds at a flow
rate of 100 µL/min.
2.4.4 Preconcentration studies
To achieve maximum immobilisation of the desired protein to the sensor surface,
preconcentration studies were performed. This important step results in an electrostatic
interaction between the negatively charged carboxyl group-rich sensor surface and
positively charged amine groups on the protein surface. At pH values lower than the
isoelectric point (pI), the protein carries a ‗net‘ positive charge and this facilitates the
binding event. The protein G-purified anti-human kappa light chain antibody diluted in
10 mM sodium acetate solution at pH values of 4.0, 4.2, 4.4, 4.6, 4.8 and 5.0, was
injected over a blank surface for 1 min at a flow rate of 5 µL/min. The pH at which the
highest ‗apparent‘ binding was observed, compared to the underivatised surface and was
used for immobilisation purposes.
2.4.5 Immobilisation conditions
The CM5 dextran surface is a three dimensional surface containing carboxyl groups
(COOH) attached to the gold surface layer. It was activated by mixing equal volumes of
400 mM EDC (N-ethyl-N-(dimethyl-aminopropyl) carbodiimide hydrochloride) and
100 mM NHS (N-hydroxysuccinimide) and injecting the mixture over the surface for 7
min at a flow rate of 10 µL/min. The EDC solution strips a H ion from the carboxyl
group resulting in a negatively charged sensor surface. The NHS solution encourages
binding of the amine groups in the protein to the COO- -rich (negatively charged)
surface. The anti-human kappa light chain, diluted in the appropriate pH buffer, was
injected over the activated surface for 30 min at 10 µL/min. Unreactived NHS groups
were capped, and deactivated and non-covalently bound proteins removed using 1 M
ethanolamine hydrochloride, pH 8.5, for 10 minutes at 10 µL/min. After capping the
surface was regenerated using 10 µL of 10 mM NaOH solution.
84
2.4.6 High throughput analysis of murine anti-fPSA scAbs using Biacore 3000
A fPSA custom kinetics wizard program was created to analyse and appropriately rank
the murine anti-fPSA scAbs selected from the library after biopanning. A capture assay
format was designed for the analysis and subsequent ranking of the murine anti-fPSA
scAbs. Each murine anti-fPSA scAb clone was grown in 20 mL 2xTY media
supplemented with 100 g/mL carbenicillin, 1 x 505 (50 % (v/v) glycerol and 5 % (v/v)
glucose solution), 1 mM MgSO4 and expressed overnight at 30oC after induction with
ITPG. The cultures were centrifuged at 4,000 rpm (3,824 x g) for 10 min at 4oC, pellets
were resuspended in 900 L PBST and the protein lysates extracted using the freezethaw method. The bacterial pellets were centrifuged at 14,000 rpm (13,384 x g) for 10
min and the lysate (900 µL) added to 100 L 10X BSA/CM dextran (120 mg/mL)
solution to reduce the chance of non-specific binding to the dextran chip surface. The
assay involved capturing three scAb clones on flow cells 2, 3 and 4 while leaving flow
cell 1 blank for reference subtraction. Each murine anti-fPSA scAb was injected over
the anti-human kappa mAb-immobilised surface for 5 min at a flow rate of 20 µL/min.
After capturing 3 murine anti-fPSA scAbs, the fPSA (30 nM) was passed over the
captured scAbs for 3 minutes at a flow rate of 30 µL/min. The fPSA molecule was
allowed dissociate for 5 min over the captured scAbs and early and late stability
responses recorded. HBS was passed over the captured scAbs and the binding response
subtracted from the fPSA binding response. Thus, double subtractive referencing was
employed. The chip surface was regenerated using a 30 second pulse of 20 mM NaOH
at a flow rate 30 L/min. The percentage stability of each clone was determined by
exporting the raw data for each curve generated using BIAevaluation software and
evaluating it in Excel. Early stability report points were calculated for a 10 second
period (between 10 and 19 seconds) starting 10 seconds after the end injection of 30 nM
fPSA, while the late stability percentage report point was calculated for a 10 second
period between 291 and 300 seconds. The complex stability was determined by dividing
the stability late by the stability early. If the stability early was less than 5 RU it was
discarded. Using a 68% complex stability cut-off of 30 nM fPSA remaining after 5
minute dissociation, selected murine anti-fPSA scAbs were further analysed by kinetic
analysis.
85
2.4.7 Kinetic analysis of selected clones using Biacore 3000 instrument
The best performing murine anti-fPSA scAbs were captured on the anti-human kappa
mAb-immobilised surface and various concentrations of fPSA (1, 0.5, 0.25, 0.125,
0.063 and 0.03 g/mL) injected over the captured antibody using HBS buffer as a
negative control for reference subtraction. The data were analysed using BIAevaluation
software and the kinetic curves generated. Association and dissociation rate constants
were calculated in addition to the affinities of each selected scAb. The calculated
affinities were very similar to those determined by kinetic analysis.
2.5 Murine anti-fPSA scAb clone sequencing
The murine anti-fPSA scAbs were sent to Eurofins MWG for sequencing. The clones
were sent as agar stab cultures in 1.5 mL centrifugation tubes supplemented with
carbenicillin (100 µg/mL) and chloramphenicol (25 µg/mL) antibiotics. The DNA
sequences for each scAb was received and the amino acid sequences determined. The
heavy and light chain regions of the antibody were identified using the Kabat rules for
antibody CDR regions. The programmes used for the amino acid sequence
determination and alignment of the clones were found in the Expasy website.
Expasy translate tool: http://www.expasy.ch/tools/dna.htmL
ClustalW: http://www.ebi.ac.uk/Tools/clustalw2/index.htmL
2.6 Selection of most sensitive murine anti-fPSA scAb by ELISA
The affinities determined by Biacore were generally quite similar, with a 10-fold
difference between the best and worst analysed scAb. Therefore, to determine the most
suitable clones for assay development an ELISA was performed. The clones were
expressed and the protein extracted by freeze-thawing. A plate coated overnight with 1
µg/mL fPSA was blocked with 3% (w/v) BSA and serial dilutions of each scAb
incubated for 1 hour at 37oC. Any binding scAbs were detected using AP-labelled antihuman kappa antibody and 4pNPP as substrate. The absorbances were read at 405 nm.
Once the optimal scAb dilution was calculated, an inhibition assay was performed with
500, 333, 222, 148, 98, 66, 44, 29, 19.5, 13 and 8.67 ng/mL fPSA. These fPSA dilutions
86
were incubated with the optimal murine anti-fPSA scAb dilution for each clone for 1
hour and added to the ELISA plate. After 1 hour incubation the wells were washed and
any bound scAb detected using AP-labelled anti-human kappa antibody and 4pNPP as
substrate. The absorbances of the wells were read at 405 nm and the scAb with the
highest degree of inhibition selected for immunoassay design.
2.7 Large-scale expression of a murine anti-fPSA scAb and subsequent purification
using immobilised metal affinity chromatography (IMAC)
Centrifugation tubes containing a culture of cells which have been induced overnight
(500 mL SB media), was spun at 12,000 rpm (11,472 x g) for 10 min at 4oC. The pellets
were resuspended in 30 mL of sonication buffer (1X PBS + 0.5 M NaCl + 20 mM
imidazole). The resuspended pellets was transferred into thirty 2 mL centrifugation
tubes (thirty one mL samples) and each sample sonicated on ice using the appropriate
setting (40 amp for 45 seconds with 3 second intervals). The sonicated-scAb containing
samples were centrifuged at 14,000 rpm (13,384 x g) for 10 min at 4oC and the lysate
was passed through a 0.2 m sterile filter. A 4 mL column of nickel NTA resin slurry
(Qiagen) was prepared and equilibrated with 30 mL of running buffer (sonication buffer
+ 1% v/v Tween). The filtered lysate was passed through the column once and the flowthrough collected. The column was washed with 30 mL of running buffer to remove any
non-specifically adsorbed proteins. The recombinant antibody fragment was eluted with
20 mL of 100 mM sodium acetate solution, pH 4.4. The eluted antibody fragment was
collected in 1.5 mL centrifugation tubes (final volume 500 L in each tube) containing
50 L 10X PBS (filtered through a 0.2 m filter) and 50 L 100 mM NaOH. An
immunodot blot was performed on each eluted fraction where 2.5 L was spotted onto
Whatman chromatography paper (0.45 µm Protran BA 85 nitrocellulose). The blot was
blocked with 3% (w/v) BSA for 1 hour at room temperature on a rotation table. The
membrane was washed with PBST (x 1) and PBS (x 1) and an AP-labelled anti-human
C kappa antibody diluted in 1% (w/v) BSA in PBS was added to the blot for 1 hour at
room temperature. After washing the blot with PBST (x 3) and PBS (x 3) the liquid
substrate (BCIP) was added. The positive fractions were pooled together and added to a
spin column with a 5,000 Dalton MWCO. This column was centrifuged at 4,000 rpm
(3,824 x g) at 4oC until concentrated to a volume of 500 L. Five mL of filtered PBS
87
was added to the column. This was stored on ice overnight at 4oC. The following day
the sample was centrifuged at 4,000 rpm (13,384 x g) at 4oC until concentrated to a 500
L volume. The concentration of protein was calculated on the Nanodrop ND-1000
spectrophotometer and an SDS-PAGE and Western blot performed. The recombinant
antibody was stored in 20 L aliquots at -20oC until required.
2.8 Induced scAb cell expression profile
A culture (100 mL SB media) supplemented with 100 g/mL carbenicillin, 25 g/mL
chloramphenicol, 1 x 505 and 1 mM MgSO4 was inoculated with 500 L of the murine
anti-fPSA scAb clone B5 culture and grown until an OD600 of 0.7 was reached. A 1 mL
sample was removed from the culture and stored at -20oC. The culture was then induced
with IPTG (1 mM final concentration) and shaken overnight at 30oC. One mL samples
were removed from the culture every hour and stored at -20oC. The protein was
extracted from each sample using the freeze-thaw method (-80oC, 37oC (x3)).
Periplasmic extracts were analysed by SDS-PAGE, Western blot and ELISA.
2.9 Sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE)
The following stock solutions were prepared for gel casting: 30% (w/v) acrylamide
containing 0.8% (w/v) bis-acrylamide, 1.5 M Tris-HCl, pH 8.8, 0.4% (w/v) SDS, 0.5 M
Tris-HCl, pH 6.8, 0.4% (w/v) SDS and 10% (w/v) ammonium persulfate. The glass
plates were washed with acetone and sealed with a gasket and grips. Free radicalinduced polymerisation of the resolving gel acrylamide (Table 2.1) was catalysed by
addition of ammonium persulphate and the accelerator TEMED and the gel added to the
space between the plates and covered with a layer of ethanol. Following polymerisation
of the gel, the ethanol was removed and the stacking gel placed directly onto the
resolving gel. A plastic comb was placed in this gel creating the wells for sample
application. Once the gel had fully polymerised, the plates were then placed in an
electrophoresis chamber, the comb was removed and the chamber and wells filled with
electrophoresis buffer (25 mM Tris, 250 mM glycine (electrophoresis grade), pH 8.3
and 0.1% (w/v) SDS).
88
Table 2.1: SDS-PAGE gel reagents
Solution
Resolving Gel
Stacking Gel
30 % Acrylamide (w/v)
10 mL
1.66 mL
Distilled Water
5.0 mL
5.75 mL
Resolving Gel Buffer
5.0 mL
-
Stacking Gel Buffer
-
2.5 mL
Ammonium Persulphate
200 mL
70 mL
TEMED
20 µL
14 µL
The gel was removed from the casting tray apparatus and stained using Coomassie blue
solution. The gel was transferred to destain solution and incubated until the protein
bands were clearly visible.
2.10 Western blotting
The SDS-PAGE gel was resolved and the proteins transferred to nitrocellulose paper.
The gel, extra thick blotting paper and nitrocellulose were incubated in transfer buffer
for 20 min before the actual transfer. Extra thick blotting paper (Sigma - #P7176) was
placed on the bottom of the transfer apparatus, the nitrocellulose was carefully placed
on top of the blotting paper, the SDS-PAGE gel was added on top of the nitrocellulose
and extra thick blotting paper placed over the SDS-PAGE gel. The proteins were
transferred to the nitrocellulose using 15 V for 20 min. The nitrocellulose was carefully
removed from the apparatus and placed in 3% (w/v) BSA solution for 1 hour. The
membrane was washed and incubated with anti-human C kappa AP-labelled antibody in
1% (w/v) BSA in PBST for 1 hour. The membrane was washed and BCIP liquid
substrate carefully added to the blot. The reaction was quenched using 100 mM EDTA
to stop the development of background colour.
2.11 Avian clone analysis on Biacore A100
The avian anti-fPSA scFvs and Fabs were analysed by Dr. Paul Leonard in Uppsala,
Sweden using the A100 Biacore. An anti-Cλ surface was created on the surface of all
four flow cells using amine coupling chemistry. Ninety six anti-fPSA Fab antibodies
89
were screened with each antibody tested against three fPSA concentrations (30, 5 and 0
nM). The analysis consisted of 3 ‗start-up‘ cycles with buffer followed by 72 sample
cycles. Each analysis cycle consisted of a 5 minutes injection of antibody across the
whole flow cell. Free PSA (fPSA) was then injected for 2 minutes with a 10 minutes
dissociation time. The surface was finally regenerated with a 30 second injection
duration of 10 mM NaOH. The time required to analyse 96 anti-fPSA Fab was 26 hours
in total. All sensorgrams were reference subtracted against an unmodified CM5 surface.
The zero concentration was then subtracted from the 30 and 5 nM sensorgrams (double
referencing) to remove any systematic instrument disturbances. A 1:1 Langmuir
equation with a mass transfer term and two separate Rmax for the two surfaces was
globally fitted to the data. Twenty six series were found acceptable (classified OK) by
the software and the determined kinetic constants can be trusted. Seventeen were
classified as inspect by the software but the reasons were either a large residual segment
or low signal to noise ratios.
Since a capture system was used and the baseline drift after the capture injection was
significant a double referencing procedure was used before the % remaining calculation
was performed. The report point at 0 nM fPSA was subtracted from the same report
point collected at 30 nM fPSA. The % remaining after dissociation was calculated by
dividing the response after 5 min dissociation (stability late response) with a report
point placed immediately after the end of the sample injection (stability early response).
2.12 Epitope Mapping of murine and avian anti-fPSA recombinant antibody
fragments using the Biacore 3000 instrument
Using an anti-human C kappa mAb-immobilised CM5 dextran chip (4-3 subtractive
referencing) epitope mapping was investigated using fPSA sandwiched between a
captured murine anti-fPSA scAb and avian anti-fPSA scFv/Fabs. The purified murine
anti-fPSA scAb clone B5 was passed over the surface of the sensor chip for 6 min at a
flow rate of 5 L/min. An aliquot containing 30 nM fPSA (1 g/mL) was passed over
the captured scAb for 2 min at a flow rate of 5 µL/min. The extracted avian clone
lysates were passed over the captured fPSA molecule for 2 min at a flow rate of 5
µL/min and the data analysed for binding. The surface was regenerated using 10 mM
90
NaOH solution for 1 min using an increased flow rate of 30 µL/min. Two negative
controls were employed to ensure no non-specific binding to the anti-human C kappa
mAb-immobilised surface or to the captured scAb fragment. FPSA (30 nM) was passed
over the immobilised anti-human kappa mAb surface to ensure no non-specific binding
to the surface had occurred. The other negative control involved passing the avian clone
lysate directly over the captured murine anti-fPSA scAb with no fPSA intermediate
molecule. No non-specific binding was observed.
2.13 Biacore subtractive inhibition assay for the detection of fPSA using the purified
murine anti-fPSA scAb
After preconcentration studies, a CM5 dextran chip was immobilised with fPSA using
EDC/NHS chemistry with approximately 6,968 RU binding. Regeneration studies were
performed to ensure a reliable, reusable working surface. The murine anti-fPSA scAb
was injected over the fPSA surface with 40 regeneration cycles using 10 mM NaOH
solution. The binding signal after each injection was measured to verify a suitable and
stable working surface for assay development. Serial dilutions of the purified murine
anti-fPSA scAb (1:50, 1:100, 1:200, 1:400, 1:800, 1:1,600, 1:3,200, 1:6,400, 1:12,800,
1:25,600, 1:51,200 and 1:102,400) diluted in HBS buffer were injected over the fPSA
immobilised surface for 2 min at a flow rate of 10 µL/min. The contact time of the
murine anti-fPSA was evaluated by increasing the injection time from 2 min to 4 and 6
min, respectively. The flow rate was evaluated to determine if more antibodies could
potentially bind to the surface at a higher flow rate. The inhibition assay was performed
using 2 murine anti-fPSA scAb dilutions (1:10,000 and 1:20,000 dilution) with a
number of fPSA concentrations (1.56 g/mL, 312, 62.5, 12.5, 6.25, 2.5, 1.0 and 0.5
ng/mL). These samples were incubated for 1 hour at 37oC and injected over the surface.
A decrease in binding response was observed with increasing fPSA concentrations.
Inter-day studies were performed at both scAb dilutions with coefficients of variation
and percentages accuracies (determined for each concentration) back-calculated from
the calibration curve.
91
2.14 Murine anti-fPSA scAb inhibition ELISA
The optimal working dilution was determined by incubating serial dilutions (1:400,
1:800, 1:1,600, 1:3,200, 1:6,400, 1:12,800, 1:25,600 and 1:51,200) of the murine antifPSA in 1% (w/v) BSA PBST for 1 hour at 37oC on a pre-blocked fPSA-coated ELISA
plate. The murine anti-fPSA scAb was detected using anti-human kappa AP-labelled
antibody and 4pNPP as substrate. The plate was incubated for 30 min at 37oC and the
absorbance read at 405 nm. The inhibition assay was performed using 2 scAb dilutions
(1:10,000 and 1:20,000) with a number of fPSA concentrations (100, 66, 44, 29, 19, 13,
8.7, 5.6 and 3.9 ng/mL). These samples were incubated for 1 hour and then added to a
fPSA-coated plate which was subsequently incubated for 1 hour at 37oC. Binding was
detected using anti-human C kappa AP-labelled antibody and 4pNPP as substrate. The
absorbances of each well was determined at 405 nm. Inter-day studies were performed
at both scAb dilutions with coefficients of variation and percentages accuracies,
determined for each concentration, back-calculated from the calibration curve.
2.15 Biacore sandwich assay using commercial anti-tPSA mAb (CanAg PSA 36 mAb)
After pre-concentration studies, the anti-tPSA mAb was successfully immobilised
(approximately 17,000 RU) onto the surface of a CM5 dextran chip using EDC/NHS
chemistry. Aliquots of fPSA (10, 5, 1 g/mL, 500, 250, 125, 62.5, 31.25, 15.63 and 7.81
ng/mL) were injected over the anti-tPSA mAb-immobilised surface and the data points
exported into Excel and a graph plotted. Various dilutions of a commercial anti-fPSA
mAb (CanAg PSA 30 mAb) (1:40,000, 1:32,000, 1:16,000, 1:8,000, 1:4,000 and
1:2,000) and murine anti-fPSA scAb (1:40,000, 1:25,000, 1:20,000, 1:15,000, 1:10,000,
1:8,000, 1:4,000, 1:2,000, 1:1,000 and 1:500) in HBS buffer containing 10X CM
Dextran/BSA (120 mg/mL) were passed over 60 ng/mL captured fPSA at 20 µL/min for
4 min. Non-specific binding was observed by injecting the diluted antibodies over the
anti-tPSA mAb surface without the fPSA interface so the assay was unreliable as the
detection antibody was binding non-specifically to the immobilised anti-tPSA mAb.
92
2.16 Biotinylation of purified murine anti-fPSA scAb for sandwich assay development
The purified murine anti-fPSA scAb was biotinylated using the Pierce EZ-link
biotinylation kit. The concentration of antibody and the levels of successful biotinylated
were calculated as follows:
A 20-fold molar excess of biotin reagent was incubated with the antibody to ensure a
high ratio level of biotin per antibody molecule. The millimoles (mmol) of biotin
reagent to add to the antibody to have a 20-fold molar excess were calculated using
equation (A):
(A) 560 µL scAb x 1.6 mg scAb x 1 mmol scAb
560 µL scAb
45,000 mg scAb
x 20 mmol biotin
1 mmol scAb
= 0.000711 mmol biotin
To calculate microlitres of 10 mM biotin reagent solution to add to the scAb reaction
equation B (below) was used:
(B) 0.000711 x
557 mg
mmol biotin
x
400 L
2.2 mg
= 72 µL biotin added to 560 µL scAb
557 = molecular weight of Sulfo-NHS-LC-biotin
400 = microlitres of H2O in which 2.2 mg Sulfo-NHS-LC-biotin is dissolved to make a
10 mM solution
A vial of sulfo-NHS-LC-biotin was removed from the freezer and equilibrated at room
temperature until required. The concentration of purified scAb (PBS) was calculated
using the Nanodrop ND-1000 spectrophotometer and the scAb then placed on ice. A 10
mM sulfo-NHS-LC-biotin solution was prepared immediately before use by dissolving
2.2 mg sulfo-NHS-LC-biotin in 400 L ultra pure H2O. Twenty seven L of this
mixture was incubated with the protein for 2 hours on ice. The NHS esters react to form
stable amide bonds with primary amino groups of proteins. After an incubation period
93
of 2 hours the protein was buffer-exchanged to remove excess non-protein bound biotin.
This was performed using a ZebraTM desalt spin column. The column was initially
centrifuged at 1,000 x g for 2 minutes to remove the storage buffer. The column was
equilibrated with 2.5 mL of PBS 150 mM, pH 7.4 and centrifuged at 1000 x g for 2 min.
The flow-through was discarded and this equilibration step was repeated twice. The
biotinylated scAb sample was applied to the resin and allowed adsorb to the resin. The
column was centrifuged at 1,000 x g for 2 min and the purified protein collected in the
flow-through. This was immediately dispensed into 20 L volumes and stored at -20oC.
The degree of biotin incorporation was determined using the HABA assay test
(http://www.piercenet.com/files/1776gj4.pdf). This involved mixing the biotinylated
antibody with a mixture of HABA and avidin. Due to the higher affinity of avidin for
biotin, the HABA is displaced and this can be evaluated at 500 nm. One mg of avidin
was mixed with 60 L of 10 mM HABA and 1.94 mL of PBS. One hundred and eighty
L of this solution was dispensed into the well of a Nunc microtitre plate and the
absorbance measured at 500 nm (A500 HABA/avidin). Twenty L of the biotinylated
protein were added to the well and the solution was mixed using a plate shaker
(IKAMTS-2/4 digital-Lennox). The HABA/avidin/biotin absorbance at 500 nm was
determined. The moles of biotin per mole of protein was calculated using the Beer
Lambert Law
A=bC
where A is the absorbance at a particular wavelength-no units for absorbance
 is the absorptivity or extinction coefficient at the wavelength (). For HABA/Avidin
at 500 nm the extinction coefficient is equal to 34000 M-1cm-1.
b is the cell path length expressed in centimetres (cm). For the microtitre plate format
the path length is 0.5 cm.
C is the concentration of the protein sample expressed in molarity (= mol/L =
mmol/mL)
Other values needed to calculate moles of biotin per moles of scAb were
Concentration of scAb in the reaction expressed as mg/mL,
94
Molecular weight of the scAb in grams per mole,
Absorbance of HABA/avidin solution at 500 nm (A500 H/A)
Absorbance of HABA/Avidin/biotin solution at 500 nm (A500 H/A/B)
Calculation of mmol protein per mL
mmol protein per mL = 1.18 mg/mL
45,000
= 0.00003
The biotin was incubated with murine anti-fPSA scAb, the complex purified and the
concentration of biotin linked to the scAb determined by adding the biotinylated scAb
(B) with a HABA (H)/avidin (A) solution. Biotin has a higher affinity for avidin and the
HABA becomes displaced. This can be measured spectrophotometrically at 500 nm.
Calculation of change in absorbance after adding the biotinylated scAb to the
HABA/avidin solution:
∆ A500 = ((A500 (0.969) – (A500 (0.747))
= 0.222
Calculation of the concentration of biotin in mmol per mL of reaction mixture
mmol biotin____ =
mL reaction mixture
∆ A (0.222)
( 34,000 x 1)
= 0.00000653
Calculation of mmol of biotin per mmol of protein
=
(0.00000653) (10) (dilution factor)
0.00002622 mmol per mL protein in original sample
= 2.49 mmol biotin per mmol scAb
95
2.17 ELISA-based assay for detection of biotinylated murine anti-fPSA scAb B5
The biotinylated scAb was analysed by ELISA to determine the optimal working
dilution for an inhibition assay. The unbiotinylated scAb was used as a negative control
to ensure no non-specific binding between the secondary extrAvidin-HRP-labelled
antibody and the purified murine anti-fPSA scAb. A plate was coated with 1 g/mL
fPSA, blocked with 3% (w/v) BSA in PBS and serial dilutions (1:100, 1:200, 1:400,
1:800, 1:1600, 1:3200, 1:6400, 1:12800 and 1:51200) of the biotinylated and
unbiotinylated scAb in 1% (w/v) BSA in PBST added in duplicate to replicate wells. An
extrAvidin-HRP-labelled antibody was added and this was detected using TMB
substrate. The reaction was quenched after a 10 minute period and the absorbance read
at 450 nm.
2.18 Sandwich assay development for improved assay sensitivity
A sandwich ELISA using the purified anti-fPSA biotinylated scAb as the capture
antibody and avian anti-fPSA clone lysates (1 scFv and 4 Fab) as detection antibodies
was investigated. A plate was coated using 5 g/mL neutravidin in PBS (150 mM, pH
7.4) for 1 hour at 37oC. The wells were blocked using 3% (w/v) BSA in PBS for 1 hour
at 37oC, washed with PBST (x 1) and PBS (x 1) and incubated with biotinylated murine
anti-fPSA scAb for 1 hour at 37oC. The wells were washed with PBST (x 2) and PBS (x
2) and a number of fPSA concentrations (100, 10, 1 and 0.1 ng/mL) were added to
individual wells, in duplicate, for 1 hour at 37oC. Any binding of fPSA was detected
using avian anti-fPSA clone lysates to determine whether the selected clones bind
different epitopes on the fPSA molecule. The wells were washed again with PBST (x 3)
and PBS (x 3) and an anti-HA-HRP-labelled secondary antibody added for 1 hour at
37oC. The wells were washed with PBST (x 3) and PBS (x 3). The substrate, TMB, was
added and plates were incubated for 10 min at room temperature. The reaction was
quenched using 10% (v/v) HCl and the absorbance read at 450 nm. The avian anti-fPSA
scFv was successfully purified by IMAC and utilised in a sandwich assay format.
96
2.19 Biacore sandwich assay using a neutrAvidin-immobilised CM5 dextran surface
After pre-concentration studies the neutravidin was successfully immobilised onto the
surface of a CM5 dextran chip (24425 RU). The biotinylated murine anti-fPSA scAb
was injected over flow cell 2 and the captured levels recorded. Aliquots of fPSA (1
g/mL, 500 ng/mL, 250 ng/mL, 125 ng/mL, 62.5 ng/mL, 31.75 ng/mL, 15.75 ng/mL,
7.5 ng/mL, 3.9 ng/mL, 1.9 ng/mL and 0.8 ng/mL) in HBS buffer were injected over the
captured biotinylated scAb at a flow rate of 10 L/min for 2 min using subtractive
referencing (2-1). The purified avian anti-fPSA scFv (1:500 dilution) was injected over
the captured fPSA using a flow rate of 30 µL/min for 2 min. The surface was
regenerated using a 30 second pulse of 2.5 mM NaOH solution.
2.20 ELISA sandwich assay using a neutravidin-coated ELISA plate
A plate (Nunc Maxisorp) coated with 5 µg/mL neutrAvidin in PBS (150 mM, pH 7.4)
for 1 hour at 37oC was blocked using 3% (w/v) BSA in PBS for 1 hour at 37oC and
washed using PBST (x 1) and PBS (x 1). An aliquot containing 1 µg/mL biotinylated
scAb (1:1,000 dilution) was incubated for 1 hour at 37oC and the plate washed with
PBST (x 2) and PBS (x 2). Serial dilutions of fPSA (25, 18.75, 14, 10.5, 7.9, 5.9, 4.45,
3.33, 2.5, 1.8, 1.4, 1.0, 0.79 and 0.59 ng/mL) in 1% (w/v) BSA in PBST were incubated
for 1 hour at 37oC. After washing with PBST (x 3) and PBS (x 3), the four anti-PSA
antibodies, (i) avian anti-fPSA scFv (1:100 dilution), (ii) commercial anti-fPSA mAb
(PSA 30) (1:1,000 dilution), (iii) commercial anti-tPSA mAb (PSA 36) (1:1,000
dilution) and (iv) commercial anti-cPSA mAb (PSA 53) (1:1,000 dilution) were
incubated in individual wells for 1 hour at 37oC in 1 % (w/v) BSA in PBST. The plate
was washed with PBST (x 3) and PBS (x 3) and each antibody detected. The avian antifPSA scFv was detected using HRP-labelled anti-HA antibody (1:2,000 dilution) and
the commercial antibodies with HRP-labelled anti-mouse IgG (1:2,000 dilution)
prepared in 1% (w/v) BSA in PBST, following incubation for 1 hour at 37oC. The plate
was washed with PBST (x 3) and PBS (x 3), substrate, TMB was added and the plate
incubated for 10 min and the reaction quenched using 10% (v/v) HCl. Absorbance
values were read at 450 nm.
97
2.21 Neutravidin ELISA plate coating concentration optimisation for improved
sandwich assay performance
Previous experience within the Applied Biochemistry Group indicated that a high
coating concentration of neutravidin was required for a workable assay format.
Therefore, a plate was coated with 10 µg/mL and 5 µg/mL neutrAvidin for 1 hour at
37oC. After the plate was blocked with 3% (w/v) BSA in PBS, a 1 µg/mL solution of
biotinylated scAb was added to plate wells and the plate incubated for 1 hour at 37oC.
After washing, a range of fPSA concentrations (25, 18.75, 14, 10.5, 7.9, 5.9, 4.45, 3.33,
2.5, 1.8, 1.4, 1.0, 0.79 and 0.59 ng/mL) were incubated (in duplicate) for 1 hour at 37oC.
After washing the avian anti-fPSA scFv (1:1,000 dilution) was added for 1 hour at 37oC
and this was detected using anti-HA-HRP-labelled antibody and TMB as substrate. The
reaction was quenched using 10% (v/v) HCl and absorbance read at 450 nm. The
optimal coating concentration determined was 5 µg/mL neutravidin.
2.22 TMB incubation time optimisation
The sandwich assay was repeated using the optimised conditions, as described in
Section 2.21, and the TMB incubation time was evaluated. The substrate was incubated
for 10 and 20 min, respectively, before quenching with 10% (v/v) HCl. The absorbance
was read at 450 nm. The sensitivity of the assay was improved using a longer TMB
incubation period with no increase in background signal. Therefore, by incubating the
substrate for 20 min a more sensitive assay was achieved.
2.23 Inter-day and intra-day studies on sandwich assay for precision and accuracy
determination
Inter-day and intra-day assay studies were performed on the sandwich ELISA using the
optimised neutravidin coating concentration and biotinylated scAb dilution with a 20
min TMB incubation period. Calibration curves were generated for both assay formats
using BIAevaluation software and the data points exported. The curves were generated
in Excel and the coefficient of variation and percentages accuracies calculated for each
concentration of fPSA.
98
2.24 Determination of the limit of detection (LOD) for fPSA in samples
A sandwich ELISA was performed using 20 blank replicates (1% (w/v) BSA in PBST)
with 20 replicates of 2.5 and 1.8 ng/mL fPSA diluted in 1% (w/v) BSA PBST (150 mM,
pH 7.4), respectively. Binding of fPSA was detected using the avian anti-fPSA scFv
clone (1:500 dilution), plates were washed with PBST (x 3) and PBS (x 3), then
incubated with HRP-labelled anti-HA antibody, washed again using PBST (x 3) and
PBS (x 3), and finally detected with TMB substrate, the reaction was quenched after 20
min using 10% (v/v) HCl. The absorbance was read at 450 nm. The LOD of the assay is
defined as where one can detect with 95% confidence at least 19 out of 20 positive
samples mixed with 20 negative samples. It is the lowest concentration of analyte where
a response can be reliably distinguished from the background.
2.25 Sandwich assay in spiked female serum
Blood was kindly donated by female members within the DCU faculty and centrifuged
at 3,500 rpm for 10 min. The serum was aliquoted and stored at -20oC. An inter-day
sandwich assay using the biotinylated scAb as the capture antibody and the avian scFv
as the detection antibody was performed using fPSA-spiked female serum. A number of
concentrations of fPSA (25, 18.75, 14, 10.5, 7.9, 5.9, 4.45, 3.33, 2.5, 1.8, 1.4, 1.0, 0.79
and 0.59 ng/mL) were added to the plate. The assay was repeated over 3 days and the
coefficient of variation and percentages accuracies determined from the calibration
curve.
2.26 Patient serum sample analysis
Serum samples from patients with PCa were kindly donated by Professor William
Watson, Conway Institute, UCD, Belfield, Dublin and stored at -80oC. The individual
samples with known tPSA concentrations, but unknown fPSA concentrations, were
analysed in duplicate using the sandwich assay format. A calibration curve was
generated and the unknown concentrations back-calculated from the curve data. The
concentration of fPSA was determined for each sample analysed and compared to the
tPSA concentration for the appropriate sample.
99
2.27 Immunohistochemical analysis
2.27.1 Sample preparation for murine and avian anti-fPSA analysis (by hand)
Representative full face sections from 2 radical prostatectomy cases of PCa were
examined for PSA expression. The sections consisted of histologically benign glands
and prostate adenocarcinoma. Suitable formalin-fixed paraffin embedded blocks from
the cases were collected, sections (4 µm thick) from the formalin fixed paraffin
embedded blocks were cut using a microtome (Leica RM2135 Rotary Microtome) and
mounted onto VBS Glass Slides Plus (S21.1910.110, Leica). The sections were dried
overnight at 37°C.
2.27.2 Immunohistochemistry using avian and murine anti-fPSA recombinant antibody
fragments
Parraffin sections were deparaffinized with two 5 min incubations in clean xylene
followed by three washes with absolute ethanol and finally by distilled water. Antigen
retrieval was carried out by incubating the sections in tri-sodium citrate (10 mM), pH
6.0, for 15 min, at high power, in a microwave oven prior to immunohistochemical
staining. After incubation the slides were rinsed for 5 min (x 2) in Tris buffer saline
(TBS – 150 mM, pH 7.6) followed by endogenenous peroxidase blocking in blocking
buffer (Envision Kit (HRP), K4006, Dako) at room temperature. The sections were then
washed for 5 minutes (x 2) in TBS and incubated with primary antibody (1:100 dilution
avian anti-fPSA scFv and murine anti-fPSA scAb) for 1 hour. After rinsing in TBS for 5
min (x 3) the sections were incubated with HRP-labelled anti-HA antibody. The
sections were washed with TBS for 5 min (x 3) and incubated with DAB chromagen
substrate (Envision Kit (HRP), K4006, Dako) for 10 min at 37oC. The reaction was
quenched by immersing the sections in distilled water. The sections incubated with the
DAB chromogen (Envision Kit (HRP), K4006, Dako) were counterstained with
haematoxylin (RBA-4213-00A, Reagecon), dehydrated, and the cover slip applied using
an automated coverslip machine (Shandon Consul Coverslipper). The tissue sections
were visualised under high magnification (100 x magnification).
100
2.27.3 Protocol for immunohistochemistry for diagnostic antibody to PSA (Dako)
(automated immunohistochemistry)
Deparrafinization, antigen retrieval and immunohistochemistry were performed on the
paraffin-embedded 4 µm tissue sections using an automated IHC platform (Bond
MaxTM - Leica Microsystems). A polymer-based detection system with 3,3diaminobenzidine (DAB) as the chromogen was used for the generation of a brown
colour. The optimal antigen retrieval, concentration of primary antibody and antigen
retrieval for PSA (anti-PSA monoclonal A0562, Dako) of 1:15,000 ER 1 for 20 min
(Epitope retrieval method 2 Bond MaxTM - Leica Microsystems) were determined by
serial dilution. The maximum signal without background immunostaining using the
appropriate positive control was determined. Sections were counterstained with
haematoxylin, dehydrated, and cover slip applied. The tissue samples were visualised
under high magnification (100X magnification).
101
Chapter 3
The generation of recombinant antibodies
to human prostate-specific antigen using
avian and murine immune systems
102
3.1 Introduction
This chapter gives an insight into the generation of recombinant antibody fragments,
phage library construction and includes an in-depth overview on the process of selecting
positively binding anti-fPSA-specific antibodies. Phage display is the predominant
technique utilised in recombinant antibody selection. A phage is a virus capable of
infecting male E. coli cells and replicating without causing cell lysis. Random segments
of DNA can be inserted into the phage particle and amplified in bacterial cells.
Antibody variable genes are commonly inserted into the coat protein of the phage and
expressed on the surface of the phage after superinfection with helper phage (Oh et al.,
2007). The phage-scFv particle can be isolated using solid phase selection with the
cognate antigen. A major advantage of using phage display is the link between the
phenotype (protein) and corresponding genotype DNA (Schmitz et al., 2000) allowing
for selection of antigen-specific phage clones. The gene encoding the displayed protein
is within the phage-scFv particle and this can easily be re-infected into E. coli cells for
propagation and further affinity maturation.
Recombinant antibody technology is an alternative approach to hybridoma technology
for generating antibodies. Recombinant antibodies can be developed using many animal
sources as long as the genome in known (Schmitz et al., 2000). Chickens have become
an ideal immunological host for the generation of recombinant antibodies due to the
selection of high affinity antibodies against highly conserved mammalian proteins
(Leonard et al., 2007). Desired clones may be selected using high throughput automated
screening facilities whereby 1,000–10,000 clones or more can be rapidly identified
(Buckler et al., 2008). It is important to note that since there is less germ-line diversity
in chickens than in mice and humans, fewer primers are required for amplifying heavy
and light chains. (Barbas et al., 2001).
The major advantage of using immunised animals over the conventional naïve antibody
library is that the host animal immune system refines specificity towards the cognate
antigen. This increases the possibility of selecting high affinity antibodies from a
generated recombinant antibody fragment library due to in vivo affinity maturation
(Schmitz et al., 2000). Although naïve libraries are becoming more popular for
103
recombinant antibody selection they lack the diversity of the immune system (somatic
hypermutation) for generating a large panel of different antibodies.
Mice and chickens were immunised with the fPSA molecule according to protocols
described by Krebber et al. (1997) and Barbas et al. (2001), respectively. The animals
were sacrificed, the spleens were harvested and the ribonucleic acid (RNA) extracted
with Trizol reagent using a well established protocol (Section 2.2.3 Barbas et al., 2001).
The amplified antibody variable genes were assembled by splice by overlap extension
polymerase chain reaction (SOE-PCR Section 2.2.8) and ligated into the pAK 100
(murine scFv) and pComb3XSS (avian scFv/Fab) vectors for ‗biopanning‘ (Section
2.2.15). This is a selection process whereby antigen-specific clones are recovered and
enriched for subsequent rounds of panning.
3.2 Murine antibody library generation
The Krebber system for recombinant antibody fragment generation allows for the
reliable cloning of functionally active antibody variable domains from a secreting
hybridoma cell or spleen primed cells (lymphocytes Krebber et al., 1997). This system
was successfully employed to generate a murine recombinant antibody fragment library
(Section 2.2.12) from the spleen cells of Balb/c mice immunised with fPSA.
3.2.1 Mice immunisation with fPSA antigen
Balb/c mice were subcutaneously immunised with fPSA over a period of 3 months and
the serum extracted from the tail of the mouse analysed by ELISA to determine the
antibody titre (level of immune response Section 2.2.2). The antibody serum titre was
determined to ensure fPSA-specific antibodies were being generated and the titre was
sufficiently high for recombinant antibody library production. The specific antibody
titre obtained was 1/100,000 when screened against fPSA (Figure 3.1). The mouse was
sacrificed to harvest the B cells from the spleen for recombinant antibody library
generation and selection of positive antibodies to fPSA.
104
Figure 3.1: Titration of serum antibodies. Serum was harvested from the tail of mice immunised
with fPSA and an antibody serum titre determined by ELISA. Bound murine antibodies were
detected using alkaline phosphatase (AP)-labelled anti-mouse antibody followed by the addition
of 4pNPP substrate. The absorbance was read at 405 nm. A high signal was observed at the
1:4,000 dilution and gradually decreases with decreasing concentration of antibody. At
1:128,000 dilution the signal is twice the background and sufficiently high for recombinant
antibody library generation.
3.2.2 Primer selection for murine heavy and light chain gene amplification
The primer sets for the amplification and assembling of the mouse variable light and
variable heavy chain domains are listed in Section 2.2.5. The respective gene fragments
(VH and VL) were amplified by PCR using these primers after successful RNA
extraction and cDNA synthesis. The gene fragments were gel purified, and coupled with
a 20 amino acid glycine-serine repeat linker. The PCR amplified recombinant antibody
fragment was compatible for use with the Krebber pAK series of vectors for selection.
The gel-purified scFv fragment and pAK 100 vector were digested for 5 hours at 37 oC
using the SfiI restriction enzyme. The restricted fragments were gel-purified, quantified
105
using a Nanodrop ND-1000 spectrophotometer and ligated overnight at room
temperature using T4 DNA ligase. This enzyme catalyses the formation of covalent
phosphodiester bonds between the 3 hydroxyl end and the 5 phosphate end. The
ligated mixture was precipitated using ethanol to concentrate the ligated DNA and to
remove any contaminating salts which may interfere with the efficiency of
electroporation (arcing is a problem often associated with electroporation where the
sample is too conductive due to salt concentration and therefore for library construction
it is vitally important to remove any residual salt prior to transformation). The DNA was
transformed into electrocompetent E. coli XL-1 Blue cells, rescued and recovered on
chloramphenicol (pAK 100 vector resistance gene) supplemented agar plates. The
library size (2.02 x 106 cfu/mL) was calculated, as described in Section 2.2.12, and the
library subjected to 5 rounds of selection on fPSA-immobilised immunotubes.
Immunotubes offer high binding capacity with a large surface area for maximum
antigen immobilisation and recovery of potential clones. The phage-scFv particles were
incubated in the fPSA-coated immunotube, eluted after washing and propagated in E.
coli cells.
3.3 Isolation of RNA from murine B cells (spleen) and first-strand DNA synthesis
The spleen was removed from the immunised mouse and carefully homogenised in
Trizol reagent. The total RNA was extracted from the tissue sample, quantified using a
Nanodrop and first-strand cDNA synthesised by reverse transcription. The RNA and
cDNA were resolved on a 1% (w/v) agarose gel (Figure 3.2) to confirm the presence of
good quality templates for further gene amplification.
106
Figure 3.2: Preparation of cDNA by reverse transcription of RNA isolation from the B
lymphocytes from a mouse spleen. Lane 1, 1 kb+ DNA molecular weight marker; lane 2, the
cDNA and lane 3, the extracted RNA.
Good quality RNA and DNA were extracted from the spleen of the immunised mouse
(Figure 3.2). The cDNA template was used in the amplification of the mouse variable
heavy and light chain genes using a range of different buffers. The variable genes were
successfully amplified by PCR (Figure 3.3) and large scale amplification was
performed.
107
3.4 PCR buffer optimisation for murine variable heavy and light chain generation
Figure 3.3: VH and VL chain optimisation using different buffer compositions. The buffer
composition used for each variable chain PCR amplification was as follows: buffer A (7.5 mM
MgCl2, pH 8.5), buffer B (10 mM MgCl2, pH 8.5), buffer D (17.5 mM MgCl2, pH 8.5), buffer G
(12.5 mM MgCl2, pH 9.5). Lane 1 and 10, 1 kb+ DNA molecular weight marker; lanes 2–5: VH
chain amplification using buffers A, B, D and G, respectively, and lanes 6–9 represent VL chain
amplification using buffer A, B, D and G, respectively.
The buffer composition that yielded the greatest amplification was buffer G. The role of
MgCl2 in the PCR is to promote DNA/DNA interactions and to form complexes with
dNTPs. The gel-purified DNA encoding VH and VL fragments were linked by splice-by
overlap extension PCR (SOE-PCR) whereby the purified fragments were joined by an
oligonucleotide encoding 20 amino acid glycine-serine linker (G4S)4. A smaller linker
would cause multimerisation of the scFv and this longer linker provides sufficient
flexibility for the VH and VL domains to associate predominantly as scFv monomers.
The SOE-PCR protein product, commonly known as the scFv fragment of the IgG
molecule, is much smaller than the IgG antibody but still retains its antigen binding
specificity to the target molecule and can be exploited for immunoassay development.
108
3.5 Murine SOE-PCR of variable heavy and light chains
The gel-purified mouse variable heavy and light chain genes were used for the
generation of the 800 bp scFv by PCR amplification. The initial PCR involved
amplifying different VH and VL template ratios to generate the desired 800 bp product.
Figure 3.4: SOE-PCR optimisation of murine variable domain gene DNA. Lane 1 contains the
1 Kb+ ladder for DNA size determination. Lanes 2 – 10 represent different concentrations of
VH: VL in the SOE-PCR. Lane 2, 10 ng VH : 10 ng VL; lane 3, 20 ng VH : 20 ng VL; lane 4, 5ng
VH : 5 ng VL; lane 5, 5 ng VH: 10 ng VL; lane 6,5 ng VH : 2.5 ng VL; lane 7, 5 ng VH : 1.25 ng VL;
lane 8, 10 ng VH: 5 ng VL; lane 9, 2.5 ng VH : 5 ng VL and lane 10 contains a PCR template
ratio of 1.25 ng VH : 5 ng VL. The PCR was performed using Phusion polymerase using the
PCR conditions described in section 2.2.8.
A non-specific band of approximately 1,000 bp was amplified in addition to the 800 bp
SOE-PCR product (Figure 3.4). A number of additives have been proven to enhance
PCR amplification (Frankman et al., 1998) increasing specificity and maximising yield
of the desire product. Therefore, after comprehensively searching the literature it was
decided to evaluate two of these additives (DMSO and BSA) which could potentially
eliminate the non-specific 1,000 bp band. The PCR annealing temperatures of the
primers can have a major effect on the amplification of a product. A touchdown PCR
109
was also performed in conjunction with the addition of DMSO (5% v/v) and BSA (50
g/mL) to the PCR amplification reaction. A touchdown PCR (Don et al., 1991)
enables binding of the primers at high initial annealing temperatures without nonspecific amplification. The annealing temperature decreases after each successive cycle
but the desired template should be abundantly present and is preferentially amplified.
Fixed concentrations of 10 ng of the VH and VL chains were utilised during the PCR
(Figure 3.5).
Figure 3.5: SOE-PCR optimisation using DMSO and BSA additives with a 0.5oC primer
annealing temperature decrease after each consecutive cycle. Lane 1 contains the 1 Kb+ DNA
ladder for DNA size determination; lane 2 represents the PCR amplification product in the
presence of 5 % (v/v) DMSO; lane 3 represents the PCR amplification product in the presence
of BSA (50 µg/mL); lane 4 represents the PCR amplification product in the presence of 5 %
(v/v) DMSO and BSA (50 µg/mL); lane 5 represent the PCR amplification product without any
additives and lane 6 is a positive scFv fragment control (internal antibody fragment).
The presence of DMSO in the PCR amplification did have a positive effect on reducing
the amplification of the non-specific 1,000 bp band. Therefore, a touchdown PCR was
performed using varying concentrations of DMSO (2%, 4%, 6%, 8% and 10% v/v) in
the PCR reaction. The touchdown PCR was performed using Phusion and Red Taq
polymerase to investigate if the type of polymerase had an effect on amplification of the
110
desired product. Betaine was another additive used for the PCR amplification. However,
no improvement in amplification was observed (results not shown).
Figure 3.6: An SOE-PCR was performed with varying DMSO concentrations using Phusion
polymerase and Red Taq polymerase. Lane 1 and 8 contain the 1 Kb+ DNA ladder for size
determination; lanes 2–6 represent the PCR amplification in the presence of 2%, 4%, 6%, 8%
and 10% (v/v) DMSO using Red Taq polymerase, respectively. Lane 7 is blank. Lanes 9–13
represent the PCR amplifications in the presence of 2%, 4%, 6%, 8% and 10% (v/v) DMSO using
Phusion polymerase, respectively.
Due to the clear difference in amplified PCR product quality provided using DMSO and
the difference between Taq and Phusion polymerases (Figure 3.6) the SOE-PCR was
also evaluated with additional polymerases with proof reading capabilities. A
touchdown PCR was performed using a number of different polymerases - Red Taq,
Phusion, Vent and Pfu polymerases in the presence of a fixed concentration of DMSO
(5%). The objective was to determine if amplification could be enhanced by using a
number of high fidelity DNA polymerases (Figure 3.7).
111
Figure 3.7: Optimisation of SOE-PCR using different polymerases for the amplification of the
scFv product. Lane 1 and 8 contain the 1 Kb+ universal ladder for DNA size determination;
lane 2 represents the PCR amplification using Red Taq polymerase; lane 3 represents the PCR
amplification using Phusion polymerase; lane 4 represents the PCR amplification using Vent
polymerase in buffer G; lane 5 represents the PCR amplification using Pfu polymerase in buffer
G; lane 6 represents a positive scFv PCR amplification using Red Taq polymerase; lane 7
represents a negative control; lane 9 represents the PCR amplification using Vent polymerase
in Vent buffer and lane 10 represents the PCR amplification using Pfu polymerase using Pfu
buffer.
Figure 3.7 shows that the optimal amplification was achieved using Red Taq
polymerase (lane 2) with little non-specific amplification of the 1,000 bp product.
Therefore, the SOE-PCR was repeated using Red Taq polymerase with various DMSO
concentrations to improve amplification. The purified VH and VL fragments were
successfully ligated together using a glycine-serine linker (Figure 3.8) after considerable
optimisation to yield an 800 bp product.
112
Figure 3.8: SOE-PCR using different DMSO concentrations to optimise the product yield. A 1
Kb+ DNA molecular weight marker was added to lane 1. The final concentration of DMSO in
each PCR reaction was 3%, 5%, 8% and 12%, respectively, as represented by lanes 2,3,4 and
5. The optimal DMSO concentration for the SOE-PCR was 3% (lane 2), as the highest yielding
product was obtained.
After large-scale amplification of the desired 800 bp PCR fragment, this DNA fragment
was gel purified, precipitated in ethanol, quantified and digested using the SfiI
restriction enzyme (Section 2.2.10). The scFv fragment was ligated into a digested pAK
100 vector and then transformed into high efficiency electrocompetent E. coli XL-1
blue cells (developed in-house), rescued and the library size calculated. The mouse antifPSA scFv library size was 2.02 x 106 cfu/mL. The library size is limited by the
transformation efficiency of electrocompetent cells and the concentration of DNA
available for electroporation (Lin et al., 2002). Large libraries sizes can be generated
and transformants rescued to yield a large panel of positive clones. In contrast, 500–
2,000 clones would be expected after a typical hybridoma fusion (Schmitz et al., 2000).
PSA-specific antibody fragments were selected on fPSA immobilised on the surface of
an immunotube, as described in Section 2.2.15.
113
3.6 Selection of fPSA-specific phage-scFv particles by ‘biopanning’
Five rounds of panning were performed and input and output titres for each round of
selection determined (Table 3.1). During the selection process, the concentration of
fPSA was decreased and the number of washes were increased as described in Section
2.2.15. This increase in stringency after each successive round of panning is important
to ensure positive clones are selected and reduces the levels of clones which have no
specificity to the target antigen. After each round of selection, the infected E. coli XL-1
Blue cells were spread uniformly across the entire surface of chloramphenicolcontaining agar plates. The plating of the bio-panned library allows every clone the
opportunity to be screened. This is important as certain clones may grow more slowly
than other clones due to the antibody toxicity within the cell. Therefore, by plating the
cells on agar plates there is an increased chance of selecting high affinity antibody
fragments. This affinity selection process allows for enrichment of recombinant phagescFv particles against the immobilised fPSA (Oh et al., 2007).
Table 3.1: Selection of murine phage displaying anti-fPSA-specific antibody fragments. A
phage-titre, before and after selection on fPSA-immobilised immunotubes, expressed as colony
forming units (cfu), is shown.
Round
(cfu/mL)
Round 1 Input
4.9 x 1011
Round 1 Output
5.0 x 106
Round 2 Input
2.0 x 1012
Round 2 Output
5.2 x 107
Round 3 Input
1.3 x 1012
Round 3 Output
3.9 x 104
Round 4 Input
1.6 x 1012
Round 4 Challenge remaining bound
3.0 x 102
Round 4 Challenge dissociated
1.2 x 105
Round 5 Input
2.4 x 1012
Round 5 Output
1.3 x 106
114
3.7 Murine polyclonal phage ELISA
Five rounds of selection were successfully completed and the eluted phage from each
round analysed by polyclonal phage ELISA described fully in Section 2.2.16 (Figure
3.9). The results of the polyclonal phage ELISA give an indication whether or not
positive clones were successfully selected from the panned library. The ELISA signal
towards the antigen of interest should increase during the course of the selection process
if positive anti-fPSA phage scFv fragments were recovered from the generated mouse
anti-fPSA library. The pooled phage contains a collection of recombinant antibody
fragments with a range of affinities for fPSA and, therefore, individual clones were
analysed by ELISA to determine the binding patterns of the clones.
Figure 3.9: Phage pools from the unpanned mouse anti-fPSA scFv glycerol library stock and
the phage pool obtained after subsequent rounds of panning were tested for binding to fPSA by
ELISA. Bound phages were detected with an anti-M13-HRP-conjugated antibody and the
absorbance read at 450 nm. A challenge was performed which involves incubating the
immunotube with free antigen to remove any weak binding phage-scFv particles. High affinity
antibodies remain attached to the fPSA-coated surface and are subsequently propagated for the
next stage of selection. A significant absorbance signal increase in pan five suggests the
presence of anti-fPSA scFv-phage fragments.
115
Sixty colonies from round 4 and 5 of selection were picked from the bio-panned stock
plates, grown in sterile 96 well microtitre plates, infected with helper phage and shaken
overnight at 30oC after the addition of kanamycin. Phage were analysed for binding to
fPSA-coated ELISA plates and detected with an HRP-conjugated anti-M13 antibody.
Approximately 8% of the selected clones showed binding to fPSA coated on the surface
of an ELISA plate from Pan 4 output and 60% were positive from the colonies selected
from Pan 5 output (Figures 3.10 and 3.11).
3.8 Murine monoclonal phage ELISA
Figure 3.10: Sixty individual colonies were randomly selected and analysed by ELISA. Five of
the expressed clones showed significant binding to a fPSA-coated ELISA plate suggesting
positive selection of phage-scFv particles from the fourth round of ‘biopanning’. Two of the five
positive clones were expressed at high levels with a ‘cut-off’ of twice the background (red
dotted line).
Figure 3.10 shows the binding of individually selected phage-scFv fragments to fPSA
following four rounds of panning. Five of the sixty individual colonies selected were
116
shown to be positive for binding to fPSA by ELISA. A ‗cut-off‘ of twice the
background was selected for positive fPSA-specific clones (red dotted line). The
monoclonal phage ELISA performed on sixty colonies selected from the fifth round of
biopanning showed that 60% of the selected clones positively bound fPSA. The
expression levels of the positive phage-scFv particles were different with varying
degrees of binding to the fPSA-coated ELISA wells.
Figure 3.11: Sixty individual colonies were randomly selected and analysed by ELISA from the
fifth round of ‘biopanning’. Sixty percent of the recombinant clones showed significant binding
to a fPSA-coated ELISA plate. The expression levels were different for each selected clone with
a selected cut-off of twice the background (red dotted line).
Three individual clones were selected at random from the positive anti-fPSA scFv
monoclonal phage ELISA plates and an inhibition phage ELISA performed using 3
fPSA concentrations (5 µg/mL, 1 µg/mL and 200 ng/mL) to investigate if the selected
clones were displaced when mixed with fPSA in solution i.e. that the antibodies
demonstrated competitive inhibition (Figure 3.12).
117
Figure 3.12: A monoclonal phage inhibition ELISA was performed on 3 random murine antifPSA scFv clones. The phage were incubated for 1 hour with fPSA (5 µg/mL, 1 µg/mL and 200
ng/mL) and added to a fPSA-coated plate. Binding phage-scFv were detected using an antiM13 HRP-labelled antibody. The absorbance readings were normalised whereby the
absorbance for each fPSA concentration incubated with antibody (A) was divided by the
maximum absorbance for each selected antibody without antigen (Ao).
The three random clones were competitive at the selected concentrations of fPSA as
there was a decrease in absorbance with increasing fPSA concentration. Two of the
clones showed similar binding patterns to fPSA and could potentially be the same clone
while the third clone has a completely different binding profile. Figure 3.12 showed that
the selected antibodies were competitive inhibitors and were displayed with fPSA in
solution (Section 2.2.17).
118
3.9 Soluble expression of mouse anti-fPSA scAb fragments
Soluble expression of the clones selected for binding to fPSA from the fifth round of
biopanning involved subcloning the scFv fragment into an appropriate expression
vector. The Krebber system uses pAK 400 expression vector for soluble expression of
mouse scFv‘s. However, it has been widely accepted that murine scFv express poorly,
have problems with folding and tend to aggregate in the periplasm (Hayhurst et al.,
2003). These issues have been raised in our laboratory and after extensively reviewing
of the literature, papers were sourced where these problems were overcome by
reformatting the scFv into a scAb. A paper by Hayhurst et al., (2003) addresses the
common pitfalls with murine scFv‘s (poor expression levels and solubility). These
scientists revealed that reformatting the recombinant antibody to a scAb significantly
improves expression levels and overall stability. The pMoPAC vectors are loosely
based on the pAK 400 Krebber vector for soluble expression of mouse scFv. The vector
has an additional human constant kappa domain upstream of the scFv insert cloning
sites. This allows for expression of a scAb format and the presence of a downstream
Skp chaperone also improves scAb solubility. The pMoPAC DNA was kindly donated
by Dr. Andrew Hayhurst after setting up collaboration with his group. It was
transformed in electrocompetent E. coli XL-1 Blue cells and stored at -80oC. The scFv
was amplified by PCR from a plasmid preparation on E. coli XL-1 Blue cellscontaining biopanning output round 5 of selection. In parallel, a plasmid purification
was performed on the pMoPAC vectors. The scFv fragment was digested using SfiI
restriction enzyme and ligated into pMoPAC 16 and pMoPAC 53 restricted vectors.
These vectors express the scFv as a single chain Ab (scAb) with a human constant
kappa domain which gives extra stability to the expressed antibody fragment and
improves overall protein expression. Electrocompetent Tuner cells were prepared and
transformed with the pRARE plasmid. Following sub-cloning the scFv into the
pMoPAC vectors, electrocompetent Tuner cells containing the pRARE plasmid were
transformed with the ligation mixture and individual clones expressed. The pRARE
plasmid codes for 6 rare tRNA codons than enhance expression of eukaryotic proteins
in prokaryotic systems. Tuner cells allow uniform entry of IPTG into the cell population
and expression can be easily regulated. One hundred and ninety two clones were grown
in 96 microwell plates and analysed by direct ELISA. Sixty percent of the selected
119
clones bound to fPSA passively immobilised on the surface of the ELISA plate (Figure
3.13).
Figure 3.13 : One hundred and ninety two individual mouse anti-fPSA scAb clones were
functionally expressed in 2xTY media supplemented with 1 x 505 (0.5% (v/v) glycerol, 0.05%
(w/v) glucose final concentration) containing 100 g/mL carbenicillin and 25 µg/mL
chloramphenicol overnight at 30oC after induction with 1 mM IPTG. A 1 µg/mL fPSA -coated
ELISA plate was blocked with 200 l of 3% (w/v) BSA in PBS. The lysates were diluted 2 fold
with 1% (w/v) BSA in PBST and added to the ELISA plate. The scAb clones were detected with
an anti-human kappa light chain antibody labelled with alkaline phosphatase. The substrate,
pNPP, was added for 30 minutes at 37oC and the resulting absorbance read at 405 nm. Sixty
percent of the expressed recombinant antibody fragments successfully bound a fPSA-coated
ELISA plate. A cut-off of twice the background (red dotted line) was selected for positive antifPSA-secreting clones.
120
3.10 Differential expression profile of an anti-PSA scFv and reformatted scAb
A comparative analysis of the expression profile between the murine anti-fPSA scFv
and its corresponding scAb was performed. A scAb with high expression levels was
selected for this study and the aim was to evaluate the expression levels of the
recombinant antibody fragments (scFv and scAb) in the same cell strain (Tuner)
containing the pRARE plasmid, induced at the same time and expressed overnight at
30oC. An ELISA was performed on the extracted recombinant antibody constructs
(scFv and scAb) to evaluate the improvement in expression of the scAb over its scFv
counterpart. The recombinant antibody fragments were expressed in 3 mL 2xTY media
overnight and analysed by ELISA (Figure 3.14).
Figure 3.14: A direct ELISA was performed where a plate was coated with 1 µg/mL fPSA,
blocked with 3 % (w/v) BSA and serial dilutions (1:2, 1:4, 1:8, 1:16, 1:32, 1:64, 1:128, 1:256
and 1:512 dilution) of the antibody lysate incubated for 1 hour at 37oC. The lysate was detected
using an anti-His-HRP-labelled antibody with TMB as the substrate of choice. The wells were
read at 450 nm. The expression level of the murine anti-fPSA scAb was considerable higher
than its murine anti-fPSA scFv counterpart.
121
This result confirms the low level of expression of the murine scFv. This is mainly due
to improper folding and protein aggregation of the scFv. Importantly, the problem of
poor expression was addressed by sub-cloning the scFv into pMoPAC series of vectors
for scAb expression. The result in Figure 3.14 confirms that by altering the recombinant
antibody format, protein expression levels can be significantly enhanced. Another
important aspect is the scAb is expressed with a human constant Kappa domain which
facilitates the detection of the expressed antibody by ELISA. This was also utilised later
in Section 4.2.3 for Biacore ranking of the selected mouse clones. The key result from
Figure 3.14 is the significant improvement of expression which allows for further
analysis of selected antibody fragments.
3.11 Selection of chicken anti-fPSA scFv fragments
Another animal model used to generate recombinant antibodies to fPSA was chickens.
Chickens were immunised with fPSA and a large library generated (3.10 x 108 cfu/mL).
The objective was to compare the selected antibody fragments from the anti-fPSA
chicken and mouse library and to determine whether or not the chicken antibodies
recognised different epitopes than the mouse antibodies on the fPSA antigen and to use
them in a sandwich assay format. Chickens are a valuable source of antibodies and have
been used extensively in biotechnology for the production of polyclonal antibodies. It
has been shown that these animals can be immunised easily with a mixture of multiple
antigens and recombinant antibodies selected, using phage display, with little or no
cross-reactivity. This section describes the selection of chicken anti-fPSA scFv
fragments from a chicken anti-fPSA library.
The avian immune system is phylogenetically distant from mammals therefore making
chickens a very good source of antibodies against some highly conserved targets, like
PSA in mammals. Large diverse chicken recombinant antibody libraries have been
generated and high affinity antibodies selected through phage display (Finlay et al.,
2005). It is important to note that one chicken can be immunised with a number of
target antigens and produce high affinity antibodies to each target with little or no crossreactivity (Finlay et al., 2006). Both the VH and VL chain loci in chickens consist of
single functional V-region genes and J-region genes. The VH has an additional D
122
segment for diversification purposes. Further diversification is achieved due to the
presence of upstream pseudogenes in both the VH and VL chains. The relative ease of
amplification of VH and VL chains due to only 2 primer sets for library production is
another advantage for using this source for recombinant antibody production. The eggs
from immunised hens are a source of large quantities of polyclonal antibodies which
can be extracted and subsequently purified. IgY antibodies have important diagnostic
and therapeutic applications particularly in detecting antigens in blood and serum
(Greunke et al., 2006).
A chicken was immunised with two antigens (PSA and BNP) for recombinant antibody
library generation. The spleen and bone marrow were removed from an immunised
chicken and the RNA extracted as previously described in the mouse antibody library
generation section. After reverse transcription the cDNA template was amplified by
PCR to generate the VH and VL chain genes. These were resolved on a 1% (w/v)
agarose gel and purified by electroelution. The purified VH and VL DNA were amplified
by PCR to generate the scFv. The Fab was generated by doing two separate SOE-PCR
amplifications and performing a final overlap to link the two constructs. The variable
heavy construct was amplified using the VH and CH chain genes and the variable light
construct was amplified using the VL and CL chain genes. Both these fragments were
gel purified by electroelution and used in the final SOE-PCR to generate the Fab. The
scFv and Fab were SfiI digested with the pComb3XSS for 5 hours at 37oC. The DNA
was purified by electroelution and the DNA concentration determined. A ligation was
performed overnight and the ligation mixture was ethanol-precipitated before the
electroporation. The transformants were rescued on carbenicillin-supplemented agar
plates at 37oC.
3.12 Optimisation of avian VH and VL chain genes amplification for Fab library
construction
The avian anti-fPSA scFv and Fab libraries were generated in conjunction with Dr.
William Finlay. The generation of the avian scFv library involved amplifying the VH
and VL chain genes separately and joining the gel-purified fragments with a glycineserine linker, as described fully in the murine anti-fPSA scFv library Section 2.3. The
123
generation of a Fab library is different as the VH and VL genes are joined to their
respective CH and CL chain genes and the Fab construct is amplified by SOE-PCR
joining the heavy variable constant domain with the light variable constant domain.
A MgCl2 gradient (3.0, 2.5, 2.0 and 1.5 mM) PCR was used for the amplification of the
avian VH and VL (kappa and lambda chain) genes. The genes were resolved on a 1%
(w/v) agarose gel (Figure 3.15) and gel-purified.
Figure 3.15: Amplification of the avian variable genes using a MgCl 2 gradient. Lanes 1 and
14:1 kb+ DNA molecular weight marker; Lanes 2–5: amplification of the variable heavy chain
gene using 3.0, 2.5, 2.0 and 1.5 mM MgCl2, respectively. Lanes 6–9: amplification of the
variable light kappa chain gene using 3.0, 2.5, 2.0 and 1.5 mM MgCl 2, respectively. Lanes 1013: amplification of the variable light lambda chain gene using 3.0, 2.5, 2.0 and 1.5 mM MgCl 2,
respectively. The PCR amplification for each chain was successful with optimal concentrations
amplified using 3.0 mM MgCl2.
The first PCR overlap involved joining the variable chains with their counterpart human
constant genes to give an 800 bp product (Figure 3.16). It is important to ensure the
construct is amplified specifically as incorrect amplification would affect the final SOEPCR overlap.
124
Figure 3.16: The first SOE-PCR using the avian variable genes and the constant human chain
genes. Lane 1: 1 kb+ DNA molecular weight marker; lane 2: heavy chain construct (VH and CH),
lane 3: kappa light chain construct (VL and CL) and lane 4: lambda light chain construct (VL
and CL). The amplification of each construct was successful using the gel-purified VH and VL
fragments.
After large scale amplification these fragments were gel purified and the Fab construct
amplified by SOE-PCR. The Fab generated consisted of the heavy/kappa chain
construct and the heavy/lambda chain construct joined using a glycine-serine linker
region. These were separately resolved on a 1% (w/v) agarose gel (Figure 3.17),
purified, pooled and ligated into the pComb 3XSS vector, ethanol precipitated and
transformed in electrocompetent E. coli XL-1 Blue cells for selection.
125
Figure 3.17: The final overlap for the generation of the Fab construct. Lane 1: 1kb+ DNA
molecular weight marker; lane 2: heavy/kappa light chain Fab and lane 3: heavy/lambda chain
Fab. The Fab fragments were successfully amplified without any non-specific amplification.
3.13 Chicken anti-fPSA scFv selection on fPSA-coated ELISA wells
The avian anti-fPSA Fab library was screened and the chicken anti-fPSA scFv library
(3.10 x 108 cfu/mL) was bio-panned against fPSA immobilised on the surface of an
ELISA plate, as described in Section 2.3.4. The input and output titres were measured
and are shown in Table 3.2. It is important to ensure no cross contamination between
libraries. If a scFv-phage particle infected the generated Fab library it would be
preferentially be packaged quicker than the Fab as it is a smaller fragment and could
potentially be selected from the library over the Fab. This would render the library
useless and cause unnecessary time loss.
126
Table 3.2: Selection of avian anti-fPSA-specific antibody fragments by phage-display. Input and
output reflect the number of phage at the start of each round of biopanning and those recovered
after each subsequent selection round, expressed as colony forming units (cfu).
Round
(cfu/mL)
Round 1 Input
3.2 x 1012
Round 1 Output
5.0 x 106
Round 2 Input
3.5 x 1012
Round 2 Output
7.0 x 107
Round 3 Input
2.4 x 1011
Round 3 Output
1.5 x 107
Three rounds of selection were completed and the eluted phage from each round
analysed by polyclonal phage ELISA as shown in Figure 3.18. The avian anti-fPSA
scFv library required 3 rounds of biopanning in comparison to 5 rounds for the murine
anti-fPSA scFv library because the avian library was 100-fold bigger and provided an
increased opportunity to select positive clones in fewer rounds of selection. After
selection the resultant phage pool was tested in an ELISA to evaluate the success of the
biopanning process. To monitor the selection process, phage pools acquired in each
round of selection were tested in parallel with phage from the unpanned library. The
ELISA signal for the antigen, fPSA, should increase during the biopanning process if
specific binders have been selected.
127
Figure 3.18: The eluted phage from the chicken anti-fPSA scFv library were analysed in
triplicate by ELISA. Three rounds of selection were performed on fPSA-coated ELISA wells and
the phage pool obtained after subsequent rounds of biopanning were tested for binding to fPSA.
Any binding phage particles were detected with a HRP-labelled anti-M13 antibody.
The signal increased from the 1st to the 3rd round of selection indicated that positive
clones were recovered from the library. The next stage of the selection process involved
soluble expression of these avian anti-fPSA scFv clones. The pComb3XSS phagemid
vector contains an amber stop codon (TAG) between the scFv and g3p which is
suppressed in E. coli XL-1 Blue cells. However, if the phage is directly infected into
non-suppressor TOP10F' cells the scFv is transcribed without the g3p fusion tag due to
the amber stop codon. Therefore, E. coli TOP10F' cells were infected with the eluted
phage pool from round three of selection and spread uniformly over the surface of agar
plates supplemented with carbenicillin (phagemid antibiotic resistance gene). The E.
coli TOP10F' cells will grow on the carbenicillin agar plates only if they have been
infected with the phagemid. Uninfected TOP10F' cells will not grow as they do not have
the appropriate antibiotic resistance gene. The plates were incubated overnight at 37oC
128
and one hundred and ninety two colonies picked and functionally expressed in sterile 96
well plates. The protein was extracted using a freeze (-80oC x 3) - thaw (37oC x 3)
method and each clone analysed by ELISA (Figure 3.19) to evaluate binding to fPSA.
The ELISA result revealed forty five clones (23%) were positively binding fPSA.
3.14 Soluble expression of chicken anti-fPSA scFv and analysis by ELISA
Figure 3.19: One hundred and ninety two individual chicken anti-fPSA scFv clones were
solubly expressed and analysed by ELISA. An ELISA plate coated with 100 L of 1 g/mL fPSA
was blocked with 200 L of 3% (w/v) BSA in PBS. The lysates were diluted two fold with 1 %
BSA (w/v) in PBST and added to the ELISA plate. Any binding scFv clones were detected with a
HRP-conjugated anti-hemagglutinin (HA) antibody.
The avian Fab library was bio-panned, E. coli TOP10F' cells infected with eluted phage
and 60 individual colonies screened. The expressed clones were analysed by ELISA
129
(Figure 3.20) to determine whether the selected clones were kappa or lambda light
chain-specific.
Figure 3.20: The avian anti-fPSA Fab lysates were analysed by ELISA to determine if they were
predominantly lambda or kappa light chains. The Fabs were analysed using an anti-lambdaHRP-labelled antibody and detected with TMB substrate. The majority selected were lambda
chain-specific (4 selected clones were not lambda chain-specific).
The majority of the selected Fabs (56) were lambda chain specific while 4 clones were
kappa chain specific (results not shown) as determined by ELISA. The clones showed a
high degree of antibody expression when grown in small scale (96-well sterile plate).
The expression levels have different patterns and could potentially be different
recombinant antibody fragments so the library diversity was checked. Individual Fab
clones from the generated library (5 clones) and from the final round of biopanning (5
130
clones) were subjected to an AluI digest. This enzyme is a high frequency cutting
restriction enzyme that recognises the sequence AG/CT and generates blunt ends. The
Fabs were resolved on a 2% (w/v) agarose gel (Figure 3.21) to visualise the diversity in
the library and the similarity in the Fabs selected from panning as described in Section
2.3.4.
Figure 3.21: AluI fragment analysis of selected avian anti-fPSA Fab clones to investigate the
diversity between the clones. Lane 1: 100 bp ladder (Invitrogen); lanes 2–6: selected clones
from the library. Lanes 7–11: clones selected from the final round of selection.
The 5 analysed clones from the new library have no similar fragment patterns
confirming a diverse selection of individual clones. Two of the 5 selected Fab clones
from the final round of selection have identical DNA sequence patterns (9 and 11). This
would suggest that these clones are potentially the same antibody fragment and are
specific for a certain epitope of the fPSA molecule. The other selected clones are less
diverse than the selected clones from the original library but are different.
131
3.15 Discussion
This chapter gives a detailed description on the generation of three recombinant
antibody fragment libraries developed from the spleens of mice and chickens
immunised with fPSA. Positively-binding anti-fPSA recombinant antibodies were
selected from the generated libraries and preliminary studies by ELISA indicated a
range of positive clones.
Separate approaches were performed for the generation of the murine and avian
recombinant antibody libraries. The murine anti-fPSA scFv library was constructed
using a method described by Krebber et al. (1997). The spleen, a rich source of
antibody-producing cells (lymphocytes) was isolated from the immunised mouse after
confirming anti-fPSA-specific antibodies were present in the serum by ELISA. The
RNA was extracted and reversed transcribed to cDNA. The variable antibody genes (VH
and VL) were successfully amplified using the primer set given in Section 2.2.5.
Problems with amplifying the SOE-PCR were encountered as a 1000 bp non-specific
band was observed initially. The SOE-PCR was optimised using DMSO in the PCR mix
and by using a touchdown PCR. The DMSO inhibits secondary structure by disrupting
base pair in GC rich templates and improves the efficiency of amplification. The
touchdown PCR enables amplification of the desired product at a high initial annealing
temperature with a decrease in annealing temperature after each successive cycle. A
range of polymerases were investigated in the PCR amplification to determine the
optimal amplification. The desired single chain Fv fragments were assembled with a
glycine-serine interface in a VL-(G4S)4-VH orientation. The flexible linker region (20
amino acids) allows folding of the generated scFv. After optimisation, the SOE-PCR
product was cloned into pAK 100 Krebber phagemid (phage plasmid) for g3p fusion
protein display. The phagemid carries a plasmid origin of replication and a phage origin
of replication. After cloning E. coli XL-1 Blue electrocompetent cells were transformed
with the DNA and the library selected on agar plates supplemented with
chloramphenicol (antibiotic resistance gene). The size of the generated library was 2.2 x
106 cfu/mL. The generated library size indicates a large number of possible clones
which could potentially bind fPSA. However, these needed to be selected from the
generated murine anti-fPSA recombinant antibody library. Krebber et al. (1997) yielded
132
a library size of 106 after electroporation. The library size is dependent on the
concentration of DNA and the efficiency of the E. coli cells.
Antigen-specific antibodies were selected from the generated library by ‗biopanning‘.
After five rounds of selection on fPSA-coated immunotubes a polyclonal ELISA
revealed the presence of positively binding anti-fPSA phage-scFv clones. Krebber et al.
(1997) reported 3 rounds of selection to be optimal for the murine scFv generated
library. The murine anti-fPSA library required 5 rounds of ‗biopanning‘ for the
selection of positive phage-scFv particles. Individual colonies were analysed by
monoclonal phage ELISA and a number of clones successfully bound fPSA-coated
wells. In the laboratory problems, where poor expression levels of murine scFv were
observed and after extensively searching of the literature, Dr. Andrew Hayhurst was
contacted about the pMoPAC vectors. The plan was to investigate if improvements in
expression levels and stability could be achieved by reformatting the recombinant
antibody. The scFv was amplified by PCR from round 5 output of selection and
subcloned into the pMoPAC series of vector for soluble expression of a scAb. This
construct greatly enhanced soluble protein expression and overall protein stability. The
ligated product was transformed in E. coli Tuner cells containing the pRARE plasmid.
Tuner cells are lacZY deletion mutants of BL21 and allow uniform entry of IPTG into
the cell population ensuring homogenous protein levels of induction. The pRARE
plasmid codes for mammalian tRNAs rarely occur in E. coli yet enhance protein
expression levels in bacterial cells restricted by codon usage. Individual colonies were
analysed for expression by ELISA confirming solubly expression of murine anti-fPSA
scAb clones. A differential expression profile of an anti-fPSA scFv and reformatted
scAb was performed by ELISA to show the significant expression levels between the
two antibody formats (Section 3.10 Figure 3.14).
Chickens were immunised with fPSA and the spleen extracted for recombinant antibody
generation. Avian anti-fPSA scFv and Fab libraries were constructed using a method
described by Andris-Widhopf and co-workers (2000). A library size of 3.10 x 108
cfu/mL for the avian anti-fPSA scFv was bio-panned against fPSA passively
immobilised on the surface of ELISA wells (Section 2.3.4). Three rounds of selection
were performed and a polyclonal phage ELISA confirmed the presence of anti-fPSA133
specific antibodies. E. coli TOP10F' cells were infected with the phage and individual
clones selected and expressed in sterile 96 wells plates. Twenty three of the 192 selected
clones were positive by ELISA. The avian anti-fPSA Fab library was bio-panned,
individual colonies selected, expressed and screened by ELISA. ELISA showed that the
Fab lysates were predominantly lambda light chain domain fragments. The Fab
construct contains the human constant domain regions and these allow for extra stability
when expressed in E.coli cells. Overall, antigen-specific recombinant antibody
fragments were successfully selected from the generated immune libraries by phage
display. A large panel of soluble clones were individually expressed and analysed by
ELISA. These selected recombinant antibodies fragments were further characterised as
described in the next chapter.
Both Krebber et al. (1997) and Andris-Widhopf et al. (2000) reported the generation of
recombinant antibody libraries from different animal sources (mice and chickens,
respectively). Krebber et al. (1997) generated 2 libraries (spleen cells fused to a tumour
cell line and B cells) of 4 x 106 and 6 x 106 transformants. Three rounds of ‗biopanning‘
were required for selection of positive binders. Andris-Widhopf and colleagues
generated 3 large chicken libraries (approximately 107 transformants), 2 scFvs (long and
short glycine-serine linker for monomeric and dimeric expression, respectively) and 1
Fab. Four rounds of ‗biopanning‘ using phage display were performed for each library
and individual antibody fragments expressed and screened with a high proportion of
positive antibody fragments selected from the library. Three rounds of selection were
required for the selection of positive chicken anti-fPSA recombinant antibody
fragments. Therefore, highly sensitive recombinant antibody fragments were selected
earlier than the expected number of rounds of ‗biopanning‘. Rojas et al. (2002)
generated a PSA recombinant antibody library size of 1.51 x 105 transformants. They
used a mouse heavy variable chain with a human light chain and successfully selected 2
scFvs that recognised PSA after 5 rounds of biopanning. Both the murine and avian
anti-fPSA libraries generated in this research were 10-fold bigger than the libraries
reported by Rojas and Andris-Widhopf (2002; 2000). The library size is very important
for selection of positive antibody fragments. The larger the library size diversity the
greater the chance of recovering high affinity antibody fragments towards the antigen.
134
Chapter 4
Screening and characterisation of putative
anti-fPSA scFv and Fab antibodies
135
4.1 Introduction
This chapter describes the strategy for selecting high affinity recombinant antibody
fragments from a panel of clones using the Biacore system. Ninety four murine antifPSA scAb-secreting clones were evaluated using a custom designed wizard on the
Biacore 3000 instrument. A capture assay was employed where the individual scAbs
were captured by anti-human C kappa mAb immobilised on the sensor surface and
fPSA was subsequently passed over the captured scAb lysates. The scAbs were ranked
according to their percentage binding stability to fPSA (30 nM) after 5 min. Preliminary
kinetic analysis was performed on 11 clones with good performance characteristics and
sequencing of the individual antibodies confirmed that they had different amino acid
sequences with variation in the complementary determining region (CDR). The best
murine anti-fPSA scAb was selected on the basis of an inhibition ELISA as the
affinities of the selected antibodies were similar. The selected clone was expressed in
large quantities and successfully purified by IMAC. The selected avian anti-fPSA scFv
and Fab clones (Section 3.11, Figures 3.19 and 3.20) were subjected to further analysis
and their affinities determined by Dr. Paul Leonard on the Biacore A100 in Uppsala,
Sweden. The same capture approach for analysing the murine anti-fPSA scAb
antibodies was pursued using two different immobilised surfaces (anti-lambda (Fab)
and anti-HA (scFv)) on the Biacore A100. After pre-concentration studies an antilambda antibody was immobilised onto the surface of a CM5 dextran chip and the Fab
lysates passed over the surface. The fPSA (30 nM) molecule was passed over the
captured Fab, the percentage binding stability calculated and the clones ranked
accordingly. Ninety six clones were evaluated using both a single fPSA concentration
injection and multiple fPSA concentration injections to determine if the association and
dissociation rate constants changed significantly due to the number of fPSA injections.
The effect of temperature (37 and 25oC) on assay performance was evaluated.
Temperature can cause shifts in the binding abilities of antibody and, therefore, is an
important parameter in assay design (Zedar-Lutz et al., 1997). The avian anti-fPSA
scFvs were examined using the capture assay format on an anti-HA immobilised
surface. Nineteen of the selected clones showed high affinities towards fPSA.
136
4.2 Biacore capture assay design
A Biacore capture assay was employed to analyse the selected antibody fragments and
to rank them based on their binding abilities.
4.2.1 Preconcentration studies of anti-human C kappa mAb
Preconcentration involves determining the optimal buffer conditions for maximum
binding of the ligand to the surface of the chip (Section 2.4.4). The protein G-purified
anti-human C kappa mAb was diluted in a low ionic strength buffer (10 mM sodium
acetate) with different pH values. The principle behind using an array of pH buffers is
that the protein carries a ‗net‘ positive charge at pH values less than its isoelectric point
causing an electrostatic interaction between the negatively charged dextran surface and
the positively charged protein. This results in a higher coupling efficiency with a lower
requirement for protein. The study is performed to ensure maximum binding levels of
the protein to the chip surface thus, ensuring a reusable assay format. The anti-human C
kappa mAb diluted in 10 mM sodium acetate buffer at various pH increments was
passed over a blank CM5 dextran surface and the response recorded. The selected pH
should be the highest pH that provides adequate immobilisation levels (Murphy et al.,
2006). The maximum potential antigen binding was observed in sodium acetate buffer,
pH 4.0 (Figure 4.1), and this was used for immobilisation of anti-human C kappa mAb
to the surface of all 4 flow cells.
4.2.2 Immobilisation of anti-human C kappa mAb on to the CM5 dextran sensor chip
surface
The immobilisation of a ligand to a CM5 dextran chip involves a number of steps
described in Section 2.4.5. The chip is firstly activated by (N-ethyl-N‘(dimethylaminopropyl) carbodiimide (EDC) and (N-hydroxysuccinimide) (NHS), the
ligand is subsequently passed over the surface and the chip surface is then deactivated to
block and eliminate any potential non-specific binding (NSB) to the cell surface. The
most utilised method of immobilisation involves the activation of the dextran surface
using EDC and NHS chemistry. The EDC solution, in the presence of NHS, activates
137
the CM dextran carboxyl groups into functional ester groups. The surface NHS esters
react with primary amino groups such as lysine residues on the protein causing covalent
binding of the protein to the surface (Murphy et al., 2006). An injection time of 7
minutes for the EDC/NHS solution activates approximately 40 % of the carboxylated
groups on the chips surface. The purity of the ligand is of vital importance for
immobilisation and should be at least 95 % as this avoids the introduction of NSB or
assay interference. Any remaining available unbound functional ester groups are
deactivated by passing 1M ethanolamine hydrochloride (pH 8.5) over the chip surface
(Casper et al., 2004).
Immobilisation of the anti-human C kappa mAb to the dextran surface was performed
by passing 60 g/mL of anti-human kappa mAb in 10 mM sodium acetate buffer, pH
4.0, over EDC/NHS activated flow cells 1, 2, 3 and 4 for 30 minutes at a flow rate of 10
l/min. After 30 minutes any remaining active groups on the surface and non-covalently
bound material were deactivated and blocked using 1M ethanolamine hydrochloride
(pH 8.5) with a 10 minute injection at a flow rate of 10 l/min. The surface was
regenerated using 10 mM NaOH to remove non-covalently attached proteins from the
surface. Approximately 12,500 RU was immobilised on the dextran surface of each
flow cell (Figure 4.2). For characterisation of antibodies from clones with varying
expression levels a high capacity surface area was essential.
138
Figure 4.1: The protein G-purified anti-human C kappa mAb, diluted in 10 mM sodium acetate
solutions of pH 4.0, 4.2, 4.4, 4.6, 4.8 and 5.0 was passed over the surface of a CM 5 dextran
chip at a flow rate of 10 µL/min. During pre-concentrations studies there is an electrostatic
interaction between the positive charged protein (due to its charge below the isoelectric point)
and the negatively charged surface carboxylated groups. The ionic strength (150 mM) of the
HBS running buffer sufficiently removed the electrostatic charged antibody from the surface.
The optimal pH for the immobilisation of the anti-human C kappa mAb to the CM5 dextran
surface was pH 4.0 as the highest levels of apparent binding was observed.
Immobilisation of anti-human C Kappa
139
Figure 4.2: The surface of a CM5 dextran chip was activated with an EDC/NHS solution and
the anti-human C kappa mAb passed over the activated carboxyl group surface. The sample
was injected for 30 minutes until the surface was fully saturated. Binding of the ligand is via the
amino groups on the protein with the activated carboxyl groups on the dextran surface chip.
After sufficient capturing of the ligand had been achieved the remaining activated groups were
deactivated with ethanolamine-HCl solution. The surface was regenerated using a 10 mM
solution of NaOH injected over the newly covalently coupled surface. The levels of anti-human
C kappa mAb immobilised onto the surface of each flow cell were 12556, 12640, 12663 and
12467 RU for flow cells 1–4, respectively.
Ninety four scAb fragments were further evaluated after preliminary studies showed a
number of positive clones as described in Section 3.8 (Figure 3.13). These were solubly
expressed in 20 mL cultures (2xTY media) and high throughput analysis performed
using a custom wizard application. After successfully immobilising anti-human C kappa
mAb on the surface of all four flow cells of a CM 5 dextran chip (Figure 4.2), a capture
format assay was performed by passing the crude bacterial scAb protein-containing
lysate over the surface of the immobilised anti-human C kappa mAb. Subtractive
referencing was implemented whereby the scAb lysate was captured on three flow cells
(2, 3 and 4) leaving flow cell 1 as a negative reference blank with no captured scAb
140
fragment (Figure 4.3). HBS and 1 g/mL fPSA (30 nM) were individually passed over
the captured scAb fragments. The refractive indexes from flow cell 1 (blank reference)
and the binding response from HBS buffer injected over each captured scAb was
subtracted from the refractive index recorded for fPSA binding to the captured scAb.
The scAbs were ranked according to the antibody-antigen complex percentage stability
remaining after 5 minute dissociation and the fragments with the highest percentage
complex stability were subjected to preliminary kinetic analysis (Section 2.4.6).
Figure 4.3: Anti-human C kappa mAb was immobilised on the surface of a CM5 dextran sensor
chip. A high throughput capture assay format was performed where the bacterial lysatecontaining scAb fragments were passed over the surface of the immobilised anti-kappa mAb
(flow cells 2, 3 and 4). A set concentration of fPSA (30 nM) was subsequently injected over all 4
flow cells and the antibody fragments ranked (double referencing) according to their complex
stability levels.
141
4.2.3 Designed custom kinetic wizard
After consulting representatives from Biacore I designed a custom wizard for the
simultaneous analysis of recombinant antibody binding clones (Figure 4.4). The
designed wizard allowed for the quick analysis and ranking of antibody binding
fragments to fPSA.
Figure 4.4: The custom designed kinetic wizard allowed the analysis of 3 murine anti-fPSA
scAbs simultaneously using double referencing (HBS over the captured murine anti-fPSA scAb
(flow cells 2, 3 and 4) and fPSA over flow cell 1 (blank – no captured scAb)). This screen
snapshot shows the list of commands used for the assay. The command icons are arranged
across the top of the file. The written file determines the flow cells for analysis (2-1, 3-1, and 41 subtractive referencing), flow rate (50 µL/min), baseline report point before sample
application, flow path for capturing the murine anti-fPSA scAb (100 µL) at a flow rate of 20
µL/min for 5 minutes, a data report point to confirm the levels of capture, injection of 30 nM
fPSA after the 3 murine anti-fPSA scAbs were captured and report points 5 seconds and 5
minutes after dissociation of the fPSA molecule.
The extracted lysates were captured on the immobilised anti-human C kappa mAb
surface and 30 nM fPSA passed over the lysates. A report point was recorded at the start
142
of dissociation (stability early) and 5 min (stability late) after the injection and the
clones ranked on their binding abilities. The percentage stability remaining after a 5 min
dissociation period gave a good overall indication of the antibody/antigen complex with
the highest affinities (Canziani et al., 2004). The analysis of 94 murine anti-fPSA scAbs
took approximately 7 days. The data were analysed using BIAevaluation software and
the data report points exported into Excel, the graph of each binding profile for each
analysed scAb were plotted (Figure 4.5).
Figure 4.5: Ninety four scAb antibody fragments were analysed and ranked according to their
binding to a set concentration of fPSA (30 nM). Each curve represents binding of fPSA to a
captured scAb fragment. The degree of association and dissociation of fPSA to the scAb
represents the apparent binding to the antibody fragment. Reports points were recorded 5
seconds after the end of the injection and 295 seconds after injection and the percentage of
stability remaining calculated and the scAb clones ranked (Table 4.1).
Each scAb was ranked according to its percentage stability after a 5 min dissociation
period. The association and dissociation rates for each selected scAb were significantly
different (Figure 4.5). Some clones have quick on rates and quick off rates while some
clones have slow on rates and slow off rates suggesting potential differences in binding
affinities for fPSA. The results were tabulated (Table 4.1) and the percentage stability
calculated (RU).
143
Table 4.1: This table represents the calculated complex stability of each scAb fragment
analysed on the Biacore 3000. A range of stability percentage values for the scAb fragments
were recorded in Excel. Binding responses less than 5 RU indicates little or no binding to 30
nM fPSA (R<5RU) and such clones were subsequently discarded.
scAb
%
scAb
%
scAb
%
scAb
%
Clone
Stability
Clone
Stability
Clone
Stability
Clone
Stability
Number
fPSA
Number
fPSA
Number
fPSA
Number
fPSA
1
N/A
25
52.92
49
N/A
73
N/A
2
53.43
26
68.20
50
N/A
74
48.30
3
N/A
27
64.29
51
N/A
75
N/A
4
N/A
28
N/A
52
N/A
76
N/A
5
70.17
29
62.35
53
N/A
77
N/A
6
76.13
30
65.67
54
N/A
78
N/A
7
N/A
31
69.01
55
N/A
79
N/A
8
N/A
32
N/A
56
N/A
80
N/A
9
82.04
33
49.36
57
N/A
81
N/A
10
62.00
34
N/A
58
N/A
82
N/A
11
60.00
35
33.79
59
N/A
83
49.84
12
N/A
36
31.31
60
60.74
84
N/A
13
N/A
37
12.94
61
N/A
85
N/A
14
60.41
38
N/A
62
61.85
86
52.70
15
57.94
39
N/A
63
61.96
87
68.39
16
34.17
40
20.34
64
N/A
88
N/A
17
71.04
41
58.31
65
N/A
89
N/A
18
69.22
42
10.48
66
N/A
90
15.13
19
63.40
43
54.36
67
N/A
91
N/A
20
N/A
44
70.73
68
N/A
92
N/A
21
N/A
45
67.55
69
75.47
93
55.76
22
33.97
46
N/A
70
67.08
94
47.21
23
N/A
47
69.70
71
6.45
24
55.31
48
59.49
72
85.00
144
Antibodies with a 68% stability or greater were selected and subjected to preliminary
kinetic analysis. This ‗cut-off‘ was chosen to reduce the risk of potentially missing high
binder scAbs. Eleven clones with percentages stabilities above 68%, were selected and
preliminary evidence demonstrated low nanomolar to picomolar range affinities using a
1:1 Langmuir model (Table 4.2).
Table 4.2: The affinities of each murine anti-fPSA scAb analysed were determined and ranked
accordingly. The association and dissociation rate constants of each scAb fragment were
recorded and preliminary affinities calculated using BIAevaluation software. The percentage
stability after 5 minute dissociation (Table 4.1) obtained for each selected scAb is presented
with the scAb name.
Clone Name
ka (1/Ms)
kd (1/s)
KD (M)
scAb F5 (68 %)
2.88 x 106
1.04 x 10-3
3.62 x 10-10 M
scAb A1 (70 %)
2.51 x 106
1.09 x 10-3
4.33 x 10-10 M
scAb F6 (82 %)
2.51 x 106
1.09 x 10-3
4.33 x 10-10 M
scAb D8 (69 %)
3.48 x 106
1.55 x 10-3
4.46 x 10-10 M
scAb E7 (70 %)
1.17 x 106
5.99 x 10-4
5.13 x 10-10 M
scAb B6 (76 %)
1.67 x 106
9.87 x 10-4
5.91 x 10-10 M
scAb C8 (69 %)
1.55 x 106
9.71 x 10-4
6.28 x 10-10 M
scAb A8 (71 %)
2.76 x 106
2.63 x 10-3
9.58 x 10-10 M
scAb B5 (69 %)
3.08 x 104
7.04 x 10-5
2.29 x 10-9 M
scAb E4 (85 %)
3.75 x 106
1.32 x 10-3
3.51 x 10-9 M
scAb C2 (75 %)
6.70 x 106
2.67 x 10-3
3.99 x 10-9 M
145
Figure 4.6: Preliminary kinetic analysis of a crude lysate from scAb F5 using a capture
approach. A number of concentrations of fPSA were passed over the captured scAb and the
curves plotted using a drifting baseline 1:1 Langmuir model in BIAevaluation. The residual plot
shows the ‘goodness’ of the fit to the line.
The curve (Figure 4.6) was fitted using a 1:1 Langmuir model with drifting baseline
with a Chi2 value of 0.172. This value gives information about the ‗goodness‘ of the
curve fit and for good qualitative data this value should be as low as possible. The
residual plot shows the data points are distributed randomly above and below the zero
line but are within acceptable limits. Initial evidence suggests that the 11 selected clones
had similar affinities for fPSA as determined using BIAevalution software on the
Biacore 3000 (Section 2.4.7). Therefore, the most suitable clone for assay development
was chosen by inhibition ELISA based on its binding ability to compete for fPSA in
solution or fPSA adsorbed onto the surface of an ELISA well. This involved expressing
each individual scAb and analysing the crude protein lysate by ELISA to determine the
146
optimal working dilution for each fragment. Inhibition studies were performed using the
appropriate working dilution for each antibody. These dilutions were incubated with
various concentrations of fPSA for 1 hour at 37oC and added to a fPSA-coated plate.
Bound murine anti-fPSA scAb was detected using anti-human C kappa AP-labelled
antibody. The substrate, pNPP, was added for 30 minutes at 37oC and the absorbance of
the wells read at 405 nm (Figure 4.7).
Figure 4.7: An inhibition ELISA using the selected murine anti-fPSA scAbs. The murine antifPSA scAbs were incubated with a range of fPSA concentrations (500, 333, 222, 148, 98, 66, 44,
29, 19.5, 13 and 8.67 ng/mL) for 1 hour at 37oC and added to a fPSA-coated plate. Bound antifPSA scAb was detected using anti-kappa alkaline phosphatase (AP)-labelled antibody. The
substrate, pNPP, was incubated for 30 minutes at 37oC and the absorbance of the wells read at
405 nm.
The ELISA analysis showed varying degrees of binding to fPSA. The majority of the
scAbs have different binding patterns as observed by ELISA but the scAb with the best
potential for immunoassay development was scAb B5. Figure 4.7 would suggest that
scAb B5 has an IC50 value less than 10 ng/mL. This selected murine anti-fPSA scAb
was subjected to further characterisation and analysis. The amino acid sequences of
each of the selected murine anti-fPSA scAbs (A1 to E4) were determined to investigate
similarities and differences in the fragments.
147
4.2.4 Alignment of scAb clones from murine anti-fPSA library
Bacterial stocks containing the murine anti-fPSA scAb clones were sent for sequencing
to Eurofins MWG as agar stab cultures in 1.5 mL centrifugation tubes supplemented
with carbenicillin (Section 2.5). The DNA sequences were received and subsequently
translated into their amino acid sequence using ExPASy translation tool. The amino
acid sequences were aligned and the complementary determining binding (CDR)
regions of the antibody evaluated using ClustalW2 software. The CDR regions of the
antibodies were highlighted according to the Kabat scheme. The glycine-serine linker
region of the clones was highlighted in green and scAb E7 is missing a glycine residue
(indicated by a dash) in the linker region but this should have no impact on binding of
the clone. The diversity between the clones is evident from the sequence variation
between the heavy and light chains which are highlighted in red (Figure 4.8). The CDR
L2, L3, H1 have slight differences in amino acid residues while CDR H2 is highly
conserved between the analysed fragments. There is more diversity observed between
the scAbs in CDR L1 and H3. The scAbs have 2 predominant amino acid sequences in
the CDR L1 region with the exception of scAb E4. It has a hydrophobic proline amino
acid residue in the CDR L1 region which is not present in any of the other sequences.
Despite its hydrophobicity prolines are usually found on protein surfaces and play an
important role in molecular recognition. Many serine residues are observed in the CDR
L1 region and these are normally common within tight turns of proteins allowing for
flexible movement and binding in the antigen pocket. A positively charged arginine was
found in scAbs B5 and E4 which frequently plays a role in protein structure and prefers
to be on the outside of proteins. However, this residue can cause non-specific binding to
negatively charged surfaces (Birtalan et al., 2008). Asparagine was present in scAbs B6,
A1 and E7 and tends to be frequently found in the active site of the protein. ScAb E7
has an isoleucine residue which is probably buried deep within the proteins hydrophobic
core. ScAbs A1, A8 and E7 have the greatest diversity in the CDR H3 region. Ippolite
et al. (2006) revealed that 40% of the amino acid residues in the CDR H3 are tyrosine
or glycine. Interestingly, scAb A1 has 5 tyrosine residues in CDR H3 which is an
important amino acid residue in creating interactions at protein-protein interfaces and
has a dominant role in antigen recognition (Koide and Sidhu, 2009). ScAb A8 has a
proline residue at the end of the CDR H3 region while scAb E7 sequence was unclear as
148
a few amino acid residues were missing (indicated by a dash). Tyrosine in combination
with glycine and serine residues have been shown to mediate high affinity antigen
recognition (Biratalan et al., 2008) and the sequenced scAb fragments possessed an
abundant level of these residues.
CDR L1
Framework 1
F6
B6
E4
A1
A7
A8
B5
E7
-XXLHGRRTQDIVMETTQSPASLPVSLGQRATISCRASEGVDLYGYSFIQ--WYQQKTGQ
-XXLHGRRTKDIVMETTQSPASLTVSLGQRATISCRASESVDSYGNSFMETHWYQQKPGQ
XMETHGRRTQDIVMETTQSPASLPVSLGQRATISCRAREGVDLYGYSFIP--WYQQKTGQ
------------RIVITQSPASLAVSLGQRATISCRASESVDSYGNSFMETHWYQQKPGQ
-XXLHGRRTQDIVMETTQSPASLPVSLGQRATISCRASEGVDLYGYSFIQ--WYQQKTGQ
XMETHGRRTQDIVMETTQSPASLPVSLGQRATISCRASEGVDLYGYSFIQ--WYQQKTGQ
XCHRHGG-IKDIVMETTPSPASLPVSLGQRATISCRAREGVDLYGYSFIQ--WYQQKTGQ
-XXCTGWRTQDI--ELTQSPASLAVSLGQRATISCRASESVDNYGISFMETNWFQQKPGQ
work 2 CDR L2
F6
B6
E4
A1
A7
A8
B5
E7
177
179
178
168
177
178
177
174
Framework 3
YWMETHWIKQRPGQGLEWIGYINPSTGYTEYSQKFKDKATLTADKSSSTAYMETQLSSLT
YWMETHWIKQRPGQGLEWIGYINPSTGYTEYSQKFKDKATLTADKSSSTAYMETQLSSLT
YWMETHWIKQRPGQGLEWIGYINPSTGYTEYSQKFKDKATLTADKSSSTALHA--LSSLT
YWMETHWIKQRPGQGLEWIGYINPSTGYTEYNQKFKDKATLTADKSSSTAYMETQLSSLT
YWMETHWIKQRPGQGLEWIGYINPSTGYTEYSQKFKDKATLTADKSSSTAYMETQLSSLT
YWMETHWIKQRPGQGLEWIGYINPSTGYTEYSQKFKDKATLTADKSSSTAYMETQLSSLT
YWMETHWIKQRPGQGLEWIGYINPSTGYTEYSQKFKDKATLTADKSSSTAYMETQLSSLT
YWMETHWIKQRPGQGLEWIGYINPSTGYTEYSQKFKDKATLTADKSSNTAYMETQLNSLT
237
239
236
228
237
238
237
234
Framework 4
CDR H3
F6
B6
E4
A1
A7
A8
B5
E7
117
119
118
108
117
118
117
117
CDR H1
KLELKRGGGGSGGGGSGGGGGSGGGGSEVQLQQSGTELAKPGASVKMETSCEASGYTFTN
KLELKRGGGGSGGGGSGGGGGSGGGGSEVQLQQSGTELAKPGASVKMETSCEASGYTFTN
KLELKRGGGGSGGGGSGGGGGSGGGGSEVQLQQSGTELAKPGASVKMETSCEASGYTFTN
KLELKRGGGGSGGGGSGGGGGSGGGGSEVQLQQSGTELAKPGASVKMETSCEASGYTFTN
KLELKRGGGGSGGGGSGGGGGSGGGGSEVQLQQSGTELAKPGASVKMETSCEASGYTFTN
KLELKRGGGGSGGGGSGGGGGSGGGGSEVQLQQSGTELAKPGASVKMETSCEASGYTFTN
KLELKRGGGGSGGGGSGGGGGSGGGGSEVQLQQSGTELAKPGASVKMETSCEASGYTFTN
KLELKRGGGGSGGGGSGGGG-SGGGGSQVQLQQSGAELARPGASVKLS--CKASGYTFTD
CDR H2
57
59
58
48
57
58
57
57
CDR L3 Frame-
Framework 1
Linker region
Framework 2
F6
B6
E4
A1
A7
A8
B5
E7
Framework 3
PPKLLISRASNLESGIPARFSGSGSRTDFTLTINPVEADDVATYYCQQSRDDPWTFGGGS
PPKLLIYRASNLESGIPARFSGSGSRTDFTLTINPVEADDVATYYCQQSRDDPWTFGGGS
PPKLLISRASNLESGIPARFSGSGSRTDFTLTINPVEADDVATYYCQQSRDDPWTFGGGS
PPKLLIYRASNLESGIPARFSGSGSRTDFTLTINPVEADDVATYYCQQSRDDPWTFGGGS
PPKLLISRASNLESGIPARFSGSGSRTDFTLTINPVEADDVATYYCQQSRDDPWTFGGGS
PPKLLISRASNLESGIPARFSGSGSRTDFTLTINPVEADDVATYYCQQSRDDPWTFGGGS
PPKLLISRASNLESGIPARFSGSGSRTDFTLTINPVEADDVATYYCQQSRDDPWTFGGGS
PPKLLIYASSNLKSGIPARFSGSGSGTDFTLNIHPVEEEDAATYYCQQSNEDPWTFGGGT
work 4
F6
B6
E4
A1
A7
A8
B5
E7
Frame-
SEDSAVYYCTRSRKLAMETDHWGQGTSVTVSS-ASGAEFAAAAPSVFIFPPSDEQLKSG 295
SEDSAVYYCTRSRKLAMETDHWGQGTSVTVSS-ASGAEFAAAAPSVFIFPPSDEQLKSG 297
SEDSAVYYCTRSRKLAMETDHWGQGTSVTVLLGLGGPNSRPPAHLSFLFPPS------- 288
SEDSAVYYCASYYGSSYATDYXGQGTSVTVSR-PRGPNSRPLHHLSSSSAH-------- 278
SEDSAVYYCTRSRKLAMETDHWGQGTSVTVSSASGAEFAAAAPSVFIFPPSDEQLKSG-- 295
SEDSAVYYCTRSRKLAMETDPWGQGNSVTVSSASGAEFAAAAPSVFTFPPSDEQL----- 293
SEDSAVYYCTRSRKLAMETDHWGQGTSVTVFSASGAEFAAAAPSVFIFPPSDEQLKSG-- 295
SEDSAVYFCAR------RLYVWGAGTTVTVSSASGAEFAAAAPSVFIFPPSDEQLKSGTASV 290
149
Figure 4.8: Murine anti-fPSA scAb gene sequences from the immunised murine anti-fPSA
library were translated into amino acid sequences using Expasy Translate tool and aligned
using ClustalW2 sequence alignment program. The highlighted sequences in red show the
CDRs (heavy and light chain) of the recombinant antibody fragments. The framework regions
are highlighted in blue and the glycine-serine linker region is green.
4.2.5 Protein expression, extraction, purification and analysis
Large-scale expression of the murine anti-fPSA scAb B5 was performed in 500 mL
culture as outlined in Section 2.7. The protein was extracted by sonication and the
bacterial pellet removed by centrifugation. The protein lysate was passed through a 0.2
µM filter to reduce clogging of the nickel resin. It was then applied to the resin and any
histidine-tagged expressed proteins captured by the nickel resin. Nickel has a high
affinity for histidine residues and other proteins pass through the column. Generally,
histidine residues are uncommon in proteins with only 2% present in globular proteins
and only half are exposed on the protein surface (Ueda et al., 2003). Material binding
non-specifically was removed by washing the resin with 20 mL of running buffer (300
mM NaCl, 20 mM imidazole and 0.5% (v/v) Tween) and the protein successfully eluted
using a low pH buffer (10 mM sodium acetate pH 4.4) (Section 2.7). An immunodot
blot visualised using an AP-labelled anti-human kappa antibody was used to confirm
the presence of the murine anti-fPSA scAb in collected fractions (Figure 4.9). The
IMAC-purified containing scAb fractions were pooled, centrifuged and analysed by
SDS-PAGE and Western blotting (Figures 4.10 and 4.11). Bacterial lysate preparation
is a critical step because the main purpose is to maximise protein extraction while
minimising unwanted protein oxidation, proteolysis and possible contamination with
genomic DNA (Structural Genomics Consortium, 2008). During IMAC purification it is
important to resuspend the protein in high ionic strength buffer to ensure protein
solubility and stability. Modest levels of imidazole in the resuspension buffer bind
proteins with less histidine residues, thus, reducing non-specific binding to the IMAC
resin (Structural Genomics Consortium, 2008).
150
Figure 4.9: Each fraction was analysed to confirm elution of the murine anti-fPSA scAb from
the nickel resin column. The individual fractions spotted (2.5 µL) on nitrocellulose paper were
unpurified lysate, flowthrough, wash and each IMAC-eluted fractions The paper was blocked
using 3% (w/v) BSA and detected using an anti-human C kappa AP-labelled antibody.
The signal intensity increases significantly after the fifth elution fraction and the
intensity of the signal is fairly uniform for all eluted fractions thereafter. The eluted
fractions were pooled and concentrated by centrifugation at 4,000 rpm at 4oC using a
5,000 molecular weight cut-off spin column. It is important to select a molecular weight
cut-off (MWCO) below (approximately 4-fold lower) the protein size to ensure
retention of sample. The protein was buffer exchanged to remove any residual sodium
acetate solution and to ensure a high protein concentration of purified sample for
analysis. Ultrafiltration is a common technique in separating proteins in size graded
fractions. The MWCO of the vivaspin column denote the size of the protein for
retention. The fractions (lysate, flow-through, wash and elution) were analysed by SDSPAGE (Figure 4.10) and Western blotting (Figure 4.11) to determine the degree of
purity and the functionality of the murine anti-fPSA scAb (Section 2.7).
151
Figure 4.10: Lane 1 and 6 represents the Sigma colorburst marker, lane 2 contains unpurified
sonicated lysate with all the proteins expressed in the bacterial cell lysate; lane 3 contains the
flow-through with all unbound proteins passing through the resin; lane 4 contains any impure
non-specific hydrophobic proteins which may weakly bind to the resin and are removed using
an increased concentration of Tween 20 and lane 5 contains the purified eluted murine antifPSA scAb with few contaminating proteins.
Figure 4.11: Lanes 1 and 6 contain the Sigma colorburst marker; lanes 2, 3 and 4 contain the
unpurified lysate, flow-through and wash fraction from the purification. Lane 5 represents the
purified protein with a molecular size of 42 kDa’s. The membrane was probed using anti-human
kappa AP-labelled antibody and developed using BCIP liquid substrate. It was quenched using
100 mM EDTA.
152
The protein samples were resolved on SDS-PAGE gels to verify that the scAb exhibited
the correct molecular weight (42 kDa) and to assess the purity of the isolated protein.
While one gel was stained with Coomassie blue to visualise individual proteins
expressed within the bacterial cell, the proteins on the other gel were transferred to a
nitrocellulose membrane and subjected to Western blot analysis using an anti-kappa
AP-labelled antibody. A large number of proteins were clearly visible in the lysate and
flow-through with little or no proteins in the wash fraction for the Coomassie bluestained SDS-PAGE gel. A single band at the correct molecular weight for the expressed
scAb fragment was evident in Figure 4.11 indicating its relative purity.
4.2.6 Time course expression profile analysis for scAb expression
An expression time course experiment was performed on the murine anti-fPSA scAb
described in Section 2.8. Before induction with IPTG (1 mM final concentration) a 1
mL sample was removed and every hour after induction 1 mL samples were removed
from the culture and stored at -20oC. In addition, a 1 mL sample was removed from the
expressed overnight culture. The samples were sonicated, the bacterial pellet
centrifuged and the protein analysed by SDS-PAGE (Figure 4.12), Western blotting
(Figure 4.13) and ELISA (Figure 4.14). The highest level of murine anti-fPSA scAb
expression was observed in the overnight sample.
153
Figure 4.12: Time course experiment for the expression of the murine anti-fPSA scAb product.
Lane 1 represents the fermentas pagerulerTM prestained marker for protein size determination
while lanes 2 – 8 represent 1 mL sample fractions harvested at 0, 1, 2, 3, 4, 5, 6 hours and
overnight post-induction, respectively.
Figure 4.13: The individual samples were analysed to investigate the levels of protein
expression over time. The proteins were transferred to a nitrocellulose membrane and
detected with an AP-labelled anti-human C kappa antibody. The level of murine antifPSA scAb increased after 3 hours of induction with the greatest level expressed in the
overnight sample.
154
Figure 4.14: The individual 1 mL samples (0, 1, 2, 3, 4, 5, 6 hours and overnight) were
analysed for expression by ELISA. The lysates were detected using an anti-human kappa APlabelled antibody. The substrate, pNPP was incubated for 30 minutes at 37oC and the
absorbances read at 405 nm.
The SDS-PAGE (Figure 4.12), Western blotting (Figure 4.13) and ELISA results
(Figure 4.14) confirmed the maximum expression levels of protein expression for the
murine anti-fPSA scAb was observed after incubating the culture overnight at 30oC
following induction with 1 mM IPTG. It is important to induce the culture at mid-to-late
log phase of growth for maximum levels of expression with a lower temperature for
induction to allow proper folding of the newly transcribed recombinant protein
(Structural genomics Consortium, 2008). They report that cells should be grown to high
cell density overnight for sufficient levels of protein expression. Large scale expression,
protein extraction and IMAC purification of the murine anti-fPSA scAb were performed
for downstream assay development.
155
4.3 Avian clone analysis
The selected avian anti-fPSA recombinant antibody fragments from panning were
subjected to further characterisation on the Biacore A100. The assay format employed
was similar to the custom design method for ranking the murine anti-fPSA scAb.
However, analysis was performed at a significantly faster rate due major advancements
in the software and hardware of the A100 compared to the Biacore 3000. The A100 is
ideal for high sample throughput analysis of proteins with high quality data analysis and
an integrated evaluation software for rapid interpretation of results.
4.3.1 Analysis of the avian anti-fPSA Fab and scFv fragments on the Biacore A100
instrument
The avian anti-fPSA scFv and Fab fragments were analysed by Dr. Paul Leonard in
Uppsala, Sweden using the A100 Biacore (Section 2.11). The avian anti-fPSA Fabs
were predominantly expressed with a constant lambda (Cλ) chain as determined by
ELISA as described in the previous, section 3.11, (Figure 3.20). Therefore, all four flow
cells were immobilised with an anti-Cλ antibody using amine coupling chemistry. The
96 avian anti-fPSA Fabs were screened against three concentrations of fPSA (30, 5 and
0 nM). The recombinant antibody was captured and fPSA injected for 2 minutes with a
10 minute dissociation time. The surface was finally regenerated with a 30 second
injection of 10 mM NaOH. The complete analysis time of the 96 anti-fPSA Fabs was 26
hours. All sensorgrams were referenced subtracted against an unmodified CM5 surface
(spot 3). The zero concentration was then subtracted from the 30 and 5 nM sensorgrams
(double referencing) to remove any systematic instrument disturbances. The data were
evaluated using a 1:1 Langmuir equation with drifting baseline and global fit. Twenty
six Fabs were classified as ‗OK‘ and therefore, the kinetic rate constant values can be
trusted. Seventeen were classified as ‗inspect‘ but the reason may be largely attributed
to low signal to noise ratio. The avian anti-fPSA scFv were analysed using a similar
capture assay approach with an anti-HA immobilised surface and kinetic analysis
performed on 7 of the 48 tested fragments.
156
The percentage stability was calculated by dividing the stability late dissociation
response (after 10 min) by the stability early response (5 seconds after the end of the
fPSA injection) and multiplying by one hundred. Subtractive referencing was
implemented whereby the refractive index from the blank flow cell and the binding
response from HBS buffer injected over the captured antibody fragments were
subtracted from the fPSA binding response
4.3.1.1 Avian anti-fPSA Fab clone analysis
The selected avian anti-fPSA Fabs were ranked according to the percentage stability of
binding to 30 nM fPSA after a 10 minute dissociation. The percentage stability for each
Fab fragment was calculated and a bar chart plotted in Excel (Figure 4.16).
Figure 4.16: Percentage stability remaining during dissociation for all Fab samples
calculated using double reference data for 30 nM on spot 1 (high immobilisation level). X-axis
scale does not contain all sample names due to space constraints. Samples that due to drift
were above 100% were set to 100%.
157
Figure 4.17: The percentage stability remaining plotted against the calculated ‘off-rates’. The
two calculated numbers show high correlation especially for the slower ‘off-rates’.
The majority of the Fabs analysed had high percentage stability after 10 minute
dissociation. The clone ranking patterns were different suggesting diverse affinities for
fPSA. An important consideration when designing the Biacore screening assay was to
determine the most productive approach that can be used for screening large panels of
antibodies in the shortest time-frame. Ideally, it would be acceptable to screen each
antibody once and determine the best clones from the data. To evaluate this, association
and dissociation rates of each clone were analysed against PSA at multiple analyte
concentrations and at 30 nM (Figure 4.18 and 4.19).
158
Figure 4.18: The association rate of nineteen of the analysed Fabs against single and multiple
fPSA concentrations were compared. The ‘on-rate’ plot shows that the majority of the Fabs
gave good correlation between single and multiple analyte analysis. However, there are Fabs,
such as 2D3, that show large variation in ‘on-rate’. In this assay setup with a very low Rmax
the system can measure ‘on-rates’ as fast as 1x107 and therefore, the data fitting for clone 2D3
is not correct and cannot be trusted. However, using multiple analyte concentrations the on-rate
is still quite fast and would be included for further analysis.
Fourteen of the 19 clones analysed had similar ‗on-rates‘ when analysed using single
and multiple injections of fPSA. Two (1A4 and 2F8) of the remaining 5 had faster ‗onrates‘ when analysing multiple injections while 3 Fabs (1D1, 2D3 and 2E7) had faster
‗on-rates‘ with a single fPSA injection.
159
Figure 4.19: The dissociation rate of the nineteen analysed Fabs against single and multiple
fPSA concentrations. It shows that the majority of these antibodies have practically the same
‘off-rate’. However, we do see some outliers with up to 10 fold difference in ‘off-rate’ for Fab
1D5. This difference is probably due to sensorgram artefacts which can occur and either
decrease or increase ‘off-rate’ values.
Sixteen of the 19 Fabs analysed had similar off rates when analysed using single and
multiple injections of fPSA. Two (1D5 and 2F8) of the remaining 3 had quicker ‗offrates‘ when analysing multiple injections while clone 1D1 had quicker off rates with a
single fPSA injection.
Kinetic analysis was performed on 26 of the best selected avian anti-fPSA Fabs. These
were selected on the basis of their percentage complex stability (Figure 4.16). The
individual Fabs were captured and then fPSA injected. The data was interpreted using
BIAevaluation software using a 1:1 Langmuir model with a drifting baseline and a
global fit. The curve was fitted with a drifting baseline as the avian anti-fPSA Fab was
dissociating from the surface and this needs to be taken into final consideration in the
evaluation of the data. A global fit was applied to the data as the response levels of the
captured avian anti-fPSA Fab lysate vary due to expression level differences.
160
Figure 4.20: The Biacore A100 instrument software comes with the added advantage of
automated data evaluation analysis allowing for quality control of data. The software ranks
the curve fitting into four categories, GOOD, ATTENTION, INSPECT and DISCARD. The
four sensorgrams shown here are examples of good fits irrespective of capture level. The
curves on the right demonstrate that the capture level is very low but the fitting is still
reasonably good.
The capture levels of fPSA are higher for Fabs 1C1 and 1C6 than 2D8 and 2D3 (Figure
4.20). However, the data curve fitting for all of the sensorgrams was acceptable. All
these clones were analysed at room temperature and temperature can have a huge effect
on binding of the antibody to the protein and this needs to be considered when
designing an immunoassay. Changes in temperature affect the binding abilities of
antibodies as well as stability. For commercialization purposes it is important to
determine which antibodies are suited for the proposed assay. The clones were analyzed
at two temperatures (25 and 37oC). Nineteen of the antibodies were classified as either
‗OK‘ or ‗inspect‘ at both temperatures. The differences in kinetics for these antibodies
were compared in two bar graphs (Figures 4.21 and 4.22)
161
Figure 4.21: Dissociation rates for the 19 antibodies with well determined kinetics at both 25
and 37oC. Most antibodies change by about a factor of 5 in ‘off-rate’. 1A5 changes very
significantly whereas 1D5 seems relatively unaffected.
Figure 4.22: Association rates for the 19 antibodies with well determined kinetics at both 25
and 37oC. Sixteen of the screened clones showed a faster on-rate at 37oC with three having a
slower association (1A4, 1D6 and 2F8).
162
Figure 4.23: ‘On/off-rate’ map of the analysed clones at 37 and 25oC. There is a general shift
to the top right corner when analysing the fragments at a higher temperature (37oC) and the
association and dissociation rates are faster but the general spread of affinities are the same.
Interestingly temperature can have a major effect on affinity as seen in clone 1B5 with an
affinity change from 170 nm at 30oC to 5 nm at 37oC. The antibodies of major interest are those
that do not change in affinity with respect to temperature change (e.g. 1D5).
Assay performance can be hugely influenced by temperature (Figure 4.23). Clones with
unfavourable shifts in performance at different temperatures (clone 1B5) are unsuitable
for certain assays. Clones (1D5) with similar affinities over a range of temperatures are
desirable for assay design.
4.3.1.2 Avian anti-fPSA scFv analysis
Forty eight avian anti-fPSA scFv were analysed using an anti-HA immobilised chip.
Four fPSA concentrations, one in duplicate and a negative (0 nM fPSA) were passed
over each captured scFv and the interaction studied (Figure 4.24). Among the 48 scFvs
only 13 showed any binding at 10 nM, which was the highest analyte concentration
included. Binding kinetics could be determined for 7 of the 13 binders. The remaining
163
scFvs showed too low binding level characteristics preventing a good assessment of the
kinetics.
Figure 4.24: Anti-HA-labelled antibody was immobilised onto 4 spots within a flow cell
using one spot as a blank reference. Two scFv fragments are captured per flow cell and
fPSA injected over every spot. The binding response from the blank flow cell and the
reference spot were subtracted from the binding response from the capture scFvs.
164
Figure 4.25: ‘On’ and ‘off-rates’ for the 7 scFv’s for which the kinetics could be determined.
The ‘off-rates’ are generally faster than for the Fab antibodies against fPSA.
Series 32 (B7), 39 (C4), 22 (D6) are the only three with affinities below 1 nM (Figure
4.25). They have calculated affinities from 50 to 500 nM. Overall, the avian anti-fPSA
recombinant antibody fragments were successfully evaluated using the Biacore A100
and avian anti-fPSA scFv clone D6 was shown to have the best affinity of all the avian
antibodies analysed.
4.4 Determination of antibody pairs for sandwich assay development
This type of assay is the one of the most common immunoassays performed in which 2
antibodies recognise different epitopes of the same antigen. The analyte is sandwiched
between the immobilised capture antibody on the surface and probed using a detection
antibody. An important consideration when designing a sandwich assay is to
categorically confirm both antibodies do not bind overlapping antigenic epitopes.
Therefore, a Biacore sandwich assay was completed to investigate antibody pairs which
165
could be utilised in a sandwich assay format. The purified murine anti-fPSA scAb was
captured using an anti-human C kappa mAb-immobilised-CM5 dextran surface chip.
The fPSA molecule (1 µg/mL) was passed over the captured antibody and probed using
avian anti-fPSA Fab and scFv lysates (Table 4.3). The assay was repeated without the
presence of the fPSA molecule to confirm no NSB between the capture and detection
antibody (Section 2.12).
Table 4.3: Epitope mapping was performed by capturing the purified murine anti-fPSA clone
B5 scAb on the anti-human kappa mAb immobilised surface, passing over 1 µg/mL fPSA and
injecting the avian clone lysates. A number of avian antibodies could be considered for a
sandwich assay format as they seem to bind different epitopes of fPSA.
Avian Clone
Captured scAb (RU)
Captured fPSA (RU)
Captured Avian (RU)
2F8 Fab
93.8
110.5
135.1
1C1 Fab
100.4
118.4
109.3
1A5 Fab
114.2
131.3
132.5
1D5 Fab
97.5
110.5
135.1
1D6 scFv
95.8
115.5
98.1
The epitope mapping Biacore assay identified a number of potential avian antibody
fragments (2F8 and 1D5 Fabs) which could be utilised in a sandwich immunoassay
format. These Fabs recognise and bind different epitopes on the fPSA molecule than the
captured murine anti-fPSA scAb.
166
4.5 Discussion
This chapter describes the screening and characterisation of anti-fPSA scAb, scFv and
Fab antibody fragments. These antibody fragments were expressed in bacteria with
different epitope tags (kappa domain for scAb, HA domain for scFv and lambda domain
for Fab) which do not interfere with the antigen/antibody binding complex formation
and seldom affect biological activity (Structural Genomics Consortium, 2008). These
tags facilitated high throughput analysis of positive clones by immobilising anti-kappa,
anti-lambda or anti-HA antibodies onto the surface of a CM5 dextran chip and
subsequent Biacore-based analysis.
In the previous chapter the selection of a large panel of positive recombinant antibody
clones from immunised animals by phage display was outlined. This created a positive
dilemma as each clone needed to be evaluated to determine the highest affinity
antibody. The analysis process can be extremely time-consuming and expensive
depending on the selection format (e.g. ELISA). Also, important information like
dynamic complex formation cannot be determined from ELISA-based assays. The
Biacore offers a rapid high-throughput ‗real-time‘ analysis format for ranking individual
clones. Leonard et al., (2007) reported screening and ranking of avian recombinant
antibody fragments using crude bacterial lysates on the Biacore A100. They developed
a capture assay format which could potentially screen 400 antibody clones per day. The
purity of the immobilising ligand is critical to prevent NSB and ensure good quality
data for interpretation. The anti-human C kappa antibody was supplied in ascites fluid
so it had to be purified using protein G, passed through a size exclusion PD-10 column
and eluted in molecular grade H2O. Preconcentration studies of the protein G-purified
anti-kappa antibody diluted in 10 mM sodium acetate buffer at different pH values
(Figure 4.1) were evaluated. The maximum levels of apparent binding were observed in
10 mM sodium acetate, pH 4.0, which was used for the immobilisation. The purified
anti-human C kappa antibody was immobilised on all four flow cell surfaces using
EDC/NHS chemistry. A custom wizard was designed on the Biacore 3000 instrument
for ranking of the murine anti-fPSA scAb clones. The wizard enabled three scAb clones
to be captured simultaneously leaving flow cell one as a blank reference. Therefore, a
capture assay format using a drifting baseline was implemented. One µg/mL fPSA (30
167
nM) was passed over the surface of all four flow cells and the percentage stability
(stability late/stability early x 100 %) for 94 antibody fragments measured over a five
minute dissociation period. The 94 fragments were ranked on the basis of their binding
to 30 nM fPSA (Table 4.1) and the best selected with a ‗cut-off‘ of over 68 % stability
after 5 minute dissociation. Kinetic analysis was performed with initial affinities
ranging from 3.99 x 10-9 M to 3.62 x 10-10 M (Table 4.2). The murine anti-fPSA B5
scAb was selected as it was found to be more inhibitive in ELISA than the other clones
even though their affinities were very similar (Figure 4.7).
The murine anti-fPSA B5 scAb was expressed in large scale (500 mL) and the protein
extracted by sonication. The protein was purified by IMAC and each eluted fraction
confirmed the presence of the murine anti-fPSA scAb by immunodot blot (Figure 4.9).
The fractions containing the purified scAb were buffer exchanged and stored at -20oC.
The fractions (lysate, flowthrough, wash and elution) were analysed by SDS-PAGE and
Western blot (Figures 4.10 and 4.11). A time course experiment confirmed that the
maximum level of expression of the murine anti-fPSA scAb was achieved overnight at
30oC. Individual samples were analysed by SDS-PAGE, Western blotting and ELISA
(Figures 4.12, 4.13 and 4.14, respectively). The murine anti-fPSA scAbs were sent to
MWG for sequencing and differences in the amino acid residues in binding regions of
the CDR confirmed the selection of a range of antibody fragments (Figure 4.8). ScAb
A1 has 3 tyrosine amino acid residues in CDR H3 which has shown to be an important
residue in protein-protein interactions. A high proportion of the tested scAb fragments
had serine residues in the CDR L1 which allows flexible movement of the antibody
fragment which is important for binding in the antigen active site. Two scAb fragments
(B5 and E5) had a positive arginine residue in the CDR L1 domain which prefers to be
positioned on the outside of the protein and is involved in protein structure. A recent
report by Sidhu et al., (2009) suggests that tyrosine, tryptophan and arginine are
common interface amino acid residues for protein-protein recognition.
The avian anti-fPSA scFvs and Fabs were analysed on the Biacore A100 instrument by
Dr. Paul Leonard in Uppsala, Sweden. They were ranked based on their percentage
stability after 10 minute dissociation. Multiple and single injection of fPSA were
examined to determine similarities or differences in binding with respect to
168
concentration. The binding abilities of the Fabs were analysed under different
temperatures (25 and 37oC) to determine the effect of temperature on each clone. The
temperature increases the association rate of 18 of the 19 clones and the dissociation
rate of 16 of the 19 analysed. The affinity of one of the avian anti-fPSA Fab (1D5)
seemed unaffected by the change in temperature, therefore making it a good antibody
candidate for assay development.
The avian anti-fPSA scFvs were captured on an anti-HA immobilised chip and
individually ranked. Out of the 48 scFvs analysed 3 scFv fragments had affinities of less
than 1 nm (B7, C4 and D6). A significant advantage of using the Biacore A100 over the
3000 instrument is the time reduction in high throughput analysis. Using the Biacore
3000 instrument, 94 murine anti-fPSA scAbs were analysed in 7 days while on the
Biacore A100 96 avian anti-fPSA Fabs were analysed in 26 hours. This is due to the
number of flow cells available for analysis. The Biacore 3000 has 4 flow cells while the
A100 has 4 flow cells with 5 spots in each flow cell allowing for 20 interactions per
cycle. The instrument has a built-in software system that selects the data and evaluates it
with confidence for good kinetic and affinity values. On the Biacore 3000 data
interpretation and analysis of results can be a very time consuming but are reliable and
reproducible. A comparative study reported by Katsamba involved 22 investigators
exploring assay reliability on the Biacore. The assay design involved immobilising the
monoclonal antibody onto the chip surface and analysing a series of PSA
concentrations. Systematic artefacts were removed from the curves for good quality data
and kinetic analysis determination using BIAevaluation software. The antibody showed
an overall affinity of 1.1 x 10-9 nM (Katsamba et al., 2006). Some of the selected
murine and avian clones show better affinity than the tested monoclonal antibody in this
study. Both the murine and avian recombinant antibody fragments have low picomolar
affinity as determined by Biacore.
Currently, few papers describing murine and avian anti-fPSA recombinant antibody
fragments have been published as most researchers have generated monoclonal
antibodies. Rojas et al. (2002) constructed a phage antibody fragment library by
combining a synthetic human light chain variable region (VL) with a diverse set of
heavy chain variable regions selected from a mouse immunised with PSA. The apparent
169
affinities for 2 clones selected were 3.5 x 10-9 M and 1.5 x 10-9 M as determined by
ELISA. Leinonen et al. (2004) investigated the development of monoclonal antibodies
against PSA by genetic immunisation. Only five monoclonal antibodies were
successfully obtained and their affinity constants were between 3.2 x 10-8 and 7.0 x 10-9
M. Popkov et al. (2004) developed rabbit monoclonal antibodies that recognise cell
surface tumour-associated antigens expressed in PCa. Their study involved using phage
display technology for the selection of human prostate cancer cell reactive antibodies,
namely, Fab antibody fragments. New Zealand white rabbits were immunised with
human prostate cancer cell lines (LNCap or DU145). Chimeric rabbit/human Fab
libraries were generated and selected by a positive/negative selection strategy. Acevedo
et al. (2002) developed a quantitative ELISA for the measurement of PSA
concentration. Mouse monoclonal antibodies with affinities of 3.7 x 10-9 M and 4.7 x
10-10 M were used to develop this assay. The assay can detect both free and complex
PSA in the sera of patients. The monoclonal antibodies discussed in this paper have
similar affinities to the generated anti-fPSA scAbs (3.99 x 10-9 M to 3.62 x 10-10 M).
However, 4 of the selected murine scAb fragments (F5, A1, F6 and D8) show better
affinity determined by Biacore. A number of the analysed avian antibody fragments
(1A5 Fab and D6 scFv) have better affinities over the reported monoclonal antibodies.
Several surface-plasmon resonance (SPR)-based assays for the detection of PSA have
been reported. Huang et al. (2005) have recently reported the development of an
immunosensor using a commercial SPR biosensor (Biacore 2000) for detection of PSA
concentrations as low as 1 ng/mL. Biacore SPR chips are usually composed of a
carboxymethyl dextran matrix linked to a gold substrate on the sensor surface.
However, self-assembled monolayers (SAMs) have been used as the immobilisation
substrate and have been shown to give better sensitivities than dextran layers (Frederix
et al., 2003). The PSA assay reported by Huang and co-workers used camel anti-PSA
antibody fragment attached to a mixed SAM of thiols for PSA capture, and a
biotinylated mouse monoclonal anti-PSA antibody and streptavidin-modified gold
nanoparticles for detection. Yu et al. (2004) reported the use of a CM5 Biacore chip as a
dextran surface onto which a sandwich assay for fPSA was located. The assay
combined SPR and fluorescent labelling, i.e. surface plasmon field-enhanced
fluorescence spectroscopy (SPFS). This technology uses surface plasmons to excite the
170
fluorescent label and produce the signal. The limit of detection observed was 80 fM for
a 40 minute contact time, many orders of magnitude lower than detection limits of PSA
assays commercially available today. The murine and avian recombinant antibody
fragments were analysed using the Biacore 3000 and A100, respectively. Capture assays
were developed with low picomolar affinity antibodies selected for both species.
Thaxton et al. (2009) reported an ultrasensitive nanoparticle-based bio-barcode assay
for the detection of PSA. The assay is approximately 300 times more sensitive than
commercially available immunoassays and has the ability to detect 330 fg/mL. It is a
promising diagnostic tool capable of detecting extremely low levels of PSA in men after
radical prostatectomy which would go unnoticed using commercial immunoassays.
Researchers are aiming to develop point-of-care testing (POCT) which will undoubtedly
help in PCa detection. Currently, the turnaround time for patient results could be a
number of weeks as samples are analysed at dedicated clinical centres. There is a
progression towards the use of label-free biosensors formats for PSA detection and the
success of these biosensor formats in future POCT devices depends on the ability to
integrate the format into devices that include simple and efficient sample handling and
readout systems (Healy et al., 2007).
171
Chapter 5
Assay development and
immunohistochemistry studies for fPSA
detection
172
5.1 Introduction
This chapter focuses primarily on the development of a sensitive assay for the detection
of fPSA by Biacore and ELISA. Inhibition and sandwich-based assays were performed
using the murine anti-fPSA scAb and avian anti-fPSA scFv. A subtractive Biacore
inhibition assay was developed using a fPSA-immobilised CM5 dextran surface chip.
The purified murine anti-fPSA scAb was incubated with varying concentrations of
fPSA, injected over the fPSA-immobilised chip surface and the binding response
recorded. A decrease in the binding levels of antibody (reported in RU) to the surface
was observed with increasing fPSA concentration. The results were normalised by
dividing the binding response for each concentration by the maximum binding response
(Ro, scAb incubated with no fPSA). The Biacore inhibition assay confirmed that the
murine anti-fPSA scAb was displaced in the presence of fPSA. Inter-day and intra-day
assays were performed to verify the accuracy and precision of the immunoassay format.
The coefficients of variation (CV‘s) were determined for each concentration and
calibration curves generated from the data using a four-parameter equation for
immunoassays using BIAevalution 3.1 softwareTM and the percentage accuracies for
each sample back-calculated from the curve. An inhibition ELISA using the murine
anti-fPSA scAb was also evaluated to compare the level of sensitivity to the Biacore
inhibition assay. Inhibition studies showed that the ELISA assay format was not as
sensitive as the Biacore assay.
The assay sensitivity was significantly improved by performing a sandwich-based
ELISA assay. The murine anti-fPSA scAb was successfully biotinylated and
subsequently captured using neutravidin with the avian anti-fPSA clones used as the
detection antibodies. The five avian (4 Fabs/1 scFv) antibody fragments were retested
by ELISA to determine whether or not they bind non-overlapping epitopes of fPSA.
Intra-day and inter-day assay calibration curves were generated with CVs and
percentage accuracies for each analyte. A Biacore-based sandwich assay was
investigated but showed no improvement in assay performance over the direct murine
anti-fPSA scAb Biacore inhibition assay. Precision assays were evaluated using 20
blanks and 20 replicates of fPSA at 2 concentrations (2.5 and 1.8 ng/mL in PBST) to
determine the limit of detection of the assay. The sandwich ELISA assay approach was
173
repeated using serum from females spiked with fPSA to determine sensitivity levels in
the serum matrix. The serum of individuals presenting with PCa and with known tPSA
values were analysed using the sandwich ELISA to determine the performance of the
assay on clinical samples. Immunohistochemistry studies were performed on prostate
tissue samples using both murine and avian anti-fPSA recombinant antibody fragments.
A key parameter for the generated recombinant antibodies is that they recognise PSA in
the context of the secreted protein in tissue samples. Therefore, studies involving
normal and PCa tissues samples were undertaken in Beaumont Hospital. It is vital that
the protein is not altered by subtle downstream changes (e.g. proteolysis) that could
minimise binding of the recombinant antibody fragments. It was confirmed that both
murine and avian anti-fPSA antibodies bound successfully to the tissue samples
analysed. However, the degree of tissue staining intensity was less than the positive
control but this was expected since the control antibody was a polyclonal and detects
tPSA while both the avian and murine recombinant antibody fragments bind fPSA
which is approximately 25 % of tPSA. This study showed that recombinant antibodies
can be successfully employed in detecting secreted proteins in prostate tissue samples
and can be used in the diagnosis of prostate cancer.
174
5.2 Development of a subtractive inhibition Biacore assay for fPSA
The purified murine anti-fPSA scAb B5 was chosen based on its overall ability to bind
fPSA in solution, as determined by ELISA in Section 4.2.4 (Figure 4.7). The fPSA
molecule was successfully immobilised onto the surface of a CM 5 dextran chip and,
following optimisation of the conditions for assay performance an inter-day study was
completed. The range of assay sensitivity as determined by Biacore was 31.3 – 6.3
ng/mL (results not shown). Biacore-based PSA assays have been reported in the
literature. Huang et al. (2005) reported a PSA SPR immunoassay using camel
antibodies immobilised on the chip surface. The concentration detected using the direct
approach was 10 ng/mL PSA. However, sensitivity was improved to the sub ng/mL by
employing a sandwich assay approach. Cuong et al. (2005) fabricated a sandwich SPR
strategy for the detection of PSA-ACT in both HBS buffer and human serum samples
with detection concentrations of 10.2 and 18.1 ng/mL, respectively.
5.3 Murine anti-fPSA inhibition ELISA assay with inter-day and intra-day studies
A solid phase inhibition-based ELISA was performed using the purified murine antifPSA scAb to compare levels of assay sensitivity with the Biacore assay format,
described in Section 5.2. After optimising the antibody working dilution for assay
development inter-day and intra-day inhibition studies were performed. Calibration
curves were generated with percentage CV and accuracy values. The range of
sensitivity for the ELISA-based inhibition assay was 100 to 3.9 ng/mL fPSA (results not
shown). This ELISA assay format showed similar levels of sensitivity as the Biacore
assay. Therefore, the assay format was modified to investigate if sensitivity could be
improved.
As reported in the literature by both Huang et al. (2005) and Cuong et al. (2005) assay
sensitivity could be further improved by implementing a sandwich-based assay. They
reported using a biotinylated antibody for sandwich assay development. Therefore, the
murine anti-fPSA scAb was biotinylated and utilised in ELISA and Biacore-based
immunoassays.
175
5.4 Biotinylation of the purified murine anti-fPSA scAb and assay development
NHS-Biotin is a 244 Da water soluble vitamin which can bind primarily to lysine
residues of proteins. The murine anti-fPSA scAb was biotinylated and employed in a
lateral
flow
immunoassay.
Naturally
biotin
has
a
high
affinity
for
streptavidin/neutravidin so a sandwich assay was investigated. The sandwich assay
format involved capturing the biotinylated murine anti-fPSA scAb on a neutravidincoated plate and using an avian anti-fPSA scFv as the detection antibody. Precision
assays were performed to determine the limit of detection of the assay with 95%
confidence. Female serum samples were spiked and assay sensitivity investigated and
serum samples from patients presenting with PCa analysed. PCa tissue samples were
analysed using the generated recombinant antibody fragments to confirm specific
binding to the secreted protein.
5.4.1 Biotinylated of the murine anti-fPSA scAb
The purified murine anti-fPSA scAb was successfully biotinylated using the Pierce EZlink biotinylation kit. During the biotinylation procedure biotin is covalently attached to
the primary amines (lysines) of the recombinant antibody. The degree of biotinylation
was determined using the HABA/Avidin test, as described in section 2.16. A sandwich
assay format was investigated with the aim of improving assay sensitivity.
5.4.2 Sandwich ELISA-based assay
After confirming by ELISA the anti-fPSA secondary antibodies bound different
epitopes of the captured fPSA molecule as described in Section 2.17 (Figure 5.1), the
optimal coating concentration of neutravidin was determined (5 µg/mL) and the TMB
substrate incubation time was investigated to compare assay sensitivity between 10 and
20 min incubation periods. A twenty minute incubation period was optimal with no
increase in background signal observed.
176
Figure 5.1: Sandwich ELISA using the captured biotinylated murine anti-fPSA scAb and antitPSA mAb, anti-fPSA mAb, anti-cPSA mAb and purified anti-fPSA scFv as detection antibodies.
Each of the primary anti-PSA antibodies were detected using the appropriate enzyme-labelled
secondary antibodies.
Both commercial (anti-tPSA/fPSA mAbs) and the avian anti-fPSA scFv successfully
recognised different epitopes on the fPSA molecule than the neutravidin captured
biotinylated murine anti-fPSA scAb. The commercial anti-cPSA mAb did not bind to
the captured fPSA molecule and no background signal was observed (Ao).
5.4.3 Pre-study validation of sandwich ELISA assay
To ensure an accurate, reliable working assay format the sandwich ELISA was repeated
over three days (inter-day assay) as decribed in Section 2.23. The sample batches were
made fresh on the day of analysis. The dilutions, incubation times and washing steps
were performed accurately each day to guarantee a reproducible assay format (Figure
5.2). The % CV and % accuracies were calculated for each concentration assessed
(Table 5.1). An intra-day sandwich ELISA was performed 4 times using the same
dilution batches over a period of one day to evaluate the overall precision of the assay.
177
The data was evaluated using a 4-parameter fit equation using BIAevaluation (Table
5.2).
Figure 5.2: Inter-day assay calibration curve for the determination of fPSA in PBST. A plate
was coated with 5 g/mL neutravidin and blocked using 3% (w/v) BSA in PBS. The biotinylated
murine anti-fPSA scAb was captured and a number of fPSA concentrations (25, 18.8, 14, 10.5,
8.0, 6.0, 4.5, 3.3, 2.5, 1.8, 1.4, 1.0, 0.8 and 0.6 ng/mL) added to the plate. The purified avian
anti-fPSA scFv was detected using an anti-HA-HRP-labelled antibody. The data is a result of
three replicate measurements over a three day period. Using BIAevaluation 3.1 software the
data was fitted with a 4 parameter equation and the original data points for each concentration
added to the curve. The error bars represent the standard deviation for each concentration
analysed in triplicate over a three day interval.
178
Table 5.1: Inter-day studies using a sandwich assay format with various fPSA concentrations
and determination of the coefficient of variation.
Concentration
Back-calculated
fPSA (ng/mL)
concentrations
% CV’s
% Accuracies
(ng/mL)
25.0
25.6
6.3
97.5
18.8
18.0
4.6
104.2
14.0
14.5
5.1
96.5
10.5
9.7
9.6
108.8
8.0
7.8
2.2
102.0
6.0
6.6
5.2
89.8
4.5
4.6
3.7
97.5
3.3
3.2
6.0
103.3
2.5
2.5
9.9
99.9
1.8
1.9
16.6
93.7
1.4
1.2
33.7
115.7
1.0
0.9
26.6
102.9
0.8
0.7
50.6
106.1
0.6
0.8
64.0
88.2
The sandwich assay can detect low levels of fPSA (1.8 ng/mL) with acceptable % CV
(16.6%) and % accuracy (93.7%) values. However, at the lower fPSA concentrations
(1.4, 1.0, 0.8 and 0.6 ng/mL) the % CVs are outside the suggested 20% limit and the
assay is unreliable at these concentrations. The range of assay sensitivity is between 25
and 1.8 ng/mL fPSA.
179
Table 5.2: Intra-day studies using nine fPSA concentrations and determination of the coefficient
of variation.
Concentration
Back-calculated
fPSA (ng/mL)
concentrations
% CV’s
% Accuracies
(ng/mL)
10.5
10.5
2.9
100.4
8.0
7.9
2.6
99.6
6.0
5.9
10.6
100.3
4.5
4.5
4.7
98.5
3.3
3.2
5.4
103.9
2.5
2.6
4.5
98.0
1.8
1.9
13.6
95.8
1.4
1.5
6.5
94.4
1.0
0.7
9.5
138.0
The intra-day study suggested that the assay could detect 1.4 ng/mL as the values were
well inside the acceptable limits. At 1.0 ng/mL the percentage accuracy of 138.02
showed that the assay was unreliable at this concentration. The inter-day and intra-day
studies show the proposed assay had good precision with CV‘s ranging from 16.61% to
2.22% between 1.8 ng/mL and 25 ng/mL fPSA and 13.55% to 2.56% between 1 ng/mL
and 10.5 ng/mL fPSA, respectively. The percentage accuracies determined in both assay
formats provide an accurate assessment of the measured fPSA concentrations. Both
assays had acceptable concentration values back-calculated from the curves.
5.4.4 Determining the limit of detection (LOD) of the sandwich ELISA assay
The LOD for any assay can be regarded as the lowest concentration which can be
reliably differentiated with 95% confidence from the limit of the blank (LOB) (Section
2.24). The LOB was determine using equation 2.3.2 and a precision assay performed
with 20 positive and negative samples using the sandwich ELISA assay approach to
investigate the accuracy of the test and to verify the LOD (Figure 5.3).
180
Figure 5.3: Twenty replicates of 2.5 ng/mL (blue diamonds) and 1.8 ng/mL (red square) fPSA
were analysed with 20 blank samples to determine the lowest concentration of analyte that can
be detected with 95 % confidence. The cut-off (LOB) was determined by getting the mean and
adding 1.645 standard deviation of the 20 negative samples (red dotted line). The value 1.645
is related to Gaussian curve where 90 % of the distribution falls between -1.645 and 1.645.
The limit of detection determined using the precision assay (Figure 5.3) was 1.8 ng/mL
fPSA with 95% confidence using 20 blanks and 20 sample replicates. After successfully
determining the limit of detection of the sandwich assay spiked female serum samples
were analysed. The serum was spiked with fPSA and a sandwich ELISA performed to
investigate assay sensitivity (Figure 5.4).
181
5.4.5 Spiked female serum samples
An inter-day study using serum samples spiked with different concentrations of fPSA
was performed to evaluate assay sensitivity (Figure 5.4) over a 3 day period (Section
2.25).
Figure 5.4: An inter-day calibration curve using female spiked serum using a 4-parameter
equation fit. The decrease in assay sensitivity is due to serum matrix which interferes with assay
sensitivity. Female serum was spiked with fPSA and analysed using a sandwich assay approach
with the biotinylated murine anti-fPSA scAb as the capture antibody and the avian anti-fPSA
scFv as the detection antibody.
182
Table 5.3: Inter-day studies using female serum spiked with different fPSA concentrations and
determination of the coefficient of variation.
Concentration
Back-calculated
fPSA (ng/mL)
concentration
% CV’s
% Accuracies
(ng/mL)
25.0
25.3
8.5
99.0
18.8
18.5
12.1
101.4
14
13.5
4.2
103.6
10.5
11.2
13.5
94.0
7.9
7.7
11.4
103.0
5.9
6.4
0.8
91.9
4.5
3.8
4.5
115.8
3.3
3.5
5.6
94.0
2.5
2.9
10.9
87.1
1.8
1.4
146.7
131.4
1.4
1.0
284.8
138.7
1.0
1.0
232.7
105.3
0.8
0.9
290.2
89.4
0.6
1.0
143.3
61.9
The female serum spiked samples (Table 5.3) had range of CV‘s (0.80% to 12.08%)
between 2.5 ng/mL and 25 ng/mL fPSA while the lower concentrations had values well
outside the acceptable limit. The calculated % accuracies were 87.08% and 115.80%
between 2.5 ng/mL and 25 ng/mL fPSA. Therefore, the LOD for the sandwich assay in
spiked female serum samples is 2.5 ng/mL fPSA. At the lower fPSA concentrations
(1.8, 1.4, 1.0, 0.79 and 0.59 ng/mL), the assay is very unreliable and does not display
any accurate values. The next stage of assay validation involved determining levels of
fPSA in clinical patient serum samples presenting with PCa symptoms.
5.4.6 Real patient serum sample analysis by sandwich ELISA
Serum samples from patients presenting with PCa were kindly donated by Professor
William Watson, Conway Institute, UCD, Belfield, Dublin. These samples were stored
183
at -80oC and analysed using the sandwich assay format using the biotinylated murine
anti-fPSA scAb as the capture antibody and the avian anti-fPSA scFv as the detection
antibody. The serum samples were added in duplicate and absorbances recorded. The
unknown fPSA concentrations were back-calculated from the curve and correlated to
the known tPSA values obtained from each sample analysed (Table 5.4).
Table 5.4: Serum samples analysed by sandwich ELISA. Five patients presenting with PCa
were tested to determine the levels of fPSA in each sample and to correlate the levels of fPSA to
the tPSA serum levels analysed using the commercially available Abbott Diagnostic AxSYM test
at Beaumont Hospital.
Sample ID
Serum analysis
Calculated fPSA
tPSA
concentration
concentration
(ng/mL)
(ng/mL)
MMH/05/0121
Yes
1.8
6.0
MMH/04/0068
Yes
1.9
6.2
MMH/04/0086
Yes
1.8
6.1
MMH/05/0104
Yes
2.2
6.4
MMH/05/0111
Yes
2.5
7.0
The levels of tPSA in the patient serum samples were previously tested in Beaumont
Hospital using the Abbott Diagnostic AxSYM tPSA test. The fPSA levels were
analysed using the sandwich assay approach and the unknown concentrations evaluated
from the calibration curve. The levels of fPSA in the sample contribute to
approximately 25% of tPSA as detected by ELISA.
5.4.7 Immunohistochemistry study using purified murine anti-fPSA scAb and avian antifPSA scFv antibodies on prostate tissue samples
This study was performed on normal and cancerous prostate tissue samples and
involved detecting fPSA localised within the tissue microarrays (TMAs). The purified
recombinant antibodies were sent to Beaumont Hospital and the assay performed by
Gillian O‘Hurley on TMAs using a Bond Max Automated Machine with a standard
diagnostic commercially available antibody as an internal positive control. Both normal
184
(BPH) and tumour (PCa) prostate tissue cells were investigated using the murine and
avian recombinant antibodies and the staining intensity evaluated. Since normal
epithelial cells express PSA it was not a surprise to observe a higher staining intensity
than tumour cells as PSA is released in the bloodstream when the architecture of the
gland is disrupted and normal function is altered. The cells were visualised under high
magnification and overall both recombinant antibody fragments bound the PSA in
prostate cells albeit with less intensity compared to the positive control polyclonal
antibody (Section 2.27 Figures 5.5, 5.6 and 5.7). The polyclonal antibody is a mixture
of antibodies that recognise many epitopes of the PSA molecule whereas the
recombinant antibody fragments are monospecific and bind to a single epitope.
Therefore, this would corroborate the stronger staining intensity observed.
Figure 5.5: Normal prostate tissue cells were analysed with the purified avian anti-fPSA scFv
using a positive control. The tissue sections were incubated with HRP-labelled secondary
antibodies using DAB chromagen as substrate. The staining intensity was greater using the
positive control antibody. However, no surrounding background stromal tissue binding was
observed with the avian anti-fPSA scFv.
185
Figure 5.6: The tumour cells have a smaller lumen than normal prostate cells and are clustered
together with little differentiation. The tissue section was screened using the purified avian
anti-fPSA scFv and incubated with HRP-labelled secondary antibodies using DAB chromagen
as substrate. The staining pattern is similar to the positive control (polyclonal antibody) but
with less intensity.
Figure 5.7: The purified murine anti-fPSA scAb demonstrated excellent staining identifying
both epithelium and secretary PSA levels in the prostate gland but is not as intensity as the
positive control test antibody (polyclonal antibody). This could be attributed to the fact that the
polyclonal antibody was raised against tPSA while the recombinant antibody fragments were
against fPSA. The lumen and epithelium of the prostate gland are clearly defined and
recognisable when stained.
186
Figure 5.8: BPH glands (non-cancerous growth of prostate) were analysed using the murine
and avian anti-fPSA recombinant antibody fragments. PSA is expressed within the gland,
however, no staining was observed. The primary antibody (murine scAb or avian scFv) was not
incubated on the tissue sections and the secondary-labelled antibody did not bind nonspecifically to the tissue sample. The slight blue staining of the nuclei is due to the destaining
solution (Haematoxylin) which causes the cells to stain brown when positive.
Both primary and secondary antibodies are required for staining of the tissue samples.
Figure 5.8 demonstrates no staining due to the fact that the primary antibody (murine or
avian recombinant antibody) is omitted from the experiment.
187
5.5 Discussion
The aim of this chapter was to assess the potential of employing SPR biosensor, ELISAbased and immunohistochemical assays for the detection of fPSA. The purified
recombinant antibody fragments described in Chapter 4 were involved in immunoassay
development and patient sample analysis.
Initial studies using a Biacore-based inhibition assay showed a working range of 31.16.3 ng/mL. This result was mirrored by an ELISA-based inhibition assay with poor
reliability at low fPSA concentrations. This was a major concern as the FDA approved
range of detection for PSA is 2-4 ng/mL. Therefore, the detection levels of fPSA were
not acceptable using the inhibition-based immunoassays. Huang et al. (2005) developed
a novel commercial PSA SPR biosensor immunoassay with detection limits of 10
ng/mL using camel antibodies. The assay involved covalently attaching the recombinant
antibody onto a gold surface and injecting PSA over the surface. They noticed that the
limit of detection could be further improved to sub ng/mL range by constructing a
sandwich assay involving a biotinylated antibody. Saerens et al. (2005) reported a
Biacore sandwich PSA assay involving capturing camel single-domain antibodies and
detecting clinical levels of different PSA in 15 minutes. Acevedo et al. (2002)
demonstrated a sandwich ELISA-based immunoassay with a limit of detection of 0.1
ng/mL using mouse monoclonal antibodies. Their ELISA-based sandwich immunoassay
incorporated highly sensitive mAbs with a detection range of 25–0.12 ng/mL which was
comparable to commercially available immunoassays. This assay format was different
to the inhibition-based immunoassays. Therefore, a sandwich assay approach was
investigated.
The underlying achievement was the development of a sandwich assay format with high
levels of sensitivity. Sandwich-based immunoassays were investigated, however, no
improvement in assay sensitivity was observed over the inhibition-based assays using a
Biacore assay format. It involved capturing fPSA using a biotinylated scAb and
detecting bound fPSA with the purified avian scFv. The fPSA capture levels by the
biotinylated scAb especially at the lower fPSA concentrations were statistically
unreliable for accurate data analysis on the Biacore. A sandwich ELISA-based assay
188
demonstrated acceptable levels of sensitivity. Inter-day and intra-day studies confirmed
a reliable working assay with a limit of detect of 1.8 ng/mL (Figure 5.2). A precision
assay confirmed the detection limit with 95% confidence (Figure 5.3). Butch et al.
(2001) investigated the analytical performance of Roche total and free PSA sandwich
assays. They reported linear ranges of 0.01–103 ng/mL and 0.02–25 ng/mL,
respectively. The CV‘s for both assays were < 5% for intra-day studies with inter-day
study values of < 13.6%. The sandwich inter-day study showed CV‘s between 2.2–
16.6% (Table 5.1) at a linear range of 25–1.8 ng/mL (Figure 5.2). The intra-day study
reported CV‘s between 2.6–13.6% (Table 5.2) at a linear range of 10.5–1.4 ng/mL. The
developed sandwich assay showed acceptable levels of detection. Commercially
available immunoassays (e.g. Roche) require expensive large automated machines to
analyse samples. These assays are performed at dedicated centres throughout the world
with trained personnel with prolonged waiting times for results (Healy et al., 2007). The
aim was to develop an economical sensitive sandwich ELISA-based immunoassay
which could be executed in a diagnostic environment using standard laboratory
equipment. To the author‘s best knowledge, this is the first reported fPSA sandwich
immunoassay using murine and avian recombinant antibody fragments.
Zundel et al. (1990) reported a sandwich-based immunoassay using 2 monoclonal
antibodies with detection levels below 2 ng/mL. In their study they confirmed no PSA
in female serum and evaluated 79 clinical male samples with known PSA
concentrations. Acevedo et al. (2002) also confirmed no PSA levels in the serum of
healthy females. Therefore, spiked female samples were evaluated using the sandwich
assay format (Figure 5.4) and a shift in sensitivity to 2.5 ng/mL was observed. Serum
matrix interference can affect the binding ability of the antibodies and impact the limit
of detection. After confirming low detection levels in spiked female serum samples real
patient samples were analysed using the sandwich ELISA format. The concentrations of
the unknown samples were calculated from the generated calibration curve (Table 5.4).
The back-calculated fPSA concentrations were approximately 25% of the tPSA
concentrations as reported using the Abbott Diagnostic AxSYM tPSA test in Beaumont
Hospital. This would suggest that the assay could potentially be used for the detection
of fPSA in clinical samples.
189
Surface-enhanced Raman scattering (SERS), ‗real-time‘ Immuno-PCR and SPR
sandwich-based assay have shown pg/mL detection levels (Healy et al., 2007). The
format severely influences the performance of the assay and sensitivity can be improved
considerably. Assays such as ‗real-time‘ immuno PCR require careful sample
preparation to prevent sample contamination, dedicated equipment and data
interpretation. Sensitivity is limited by background signal due to non-specific adsorption
of assay reagents to the walls of the tube (Lind et al., 2005). There are a number of
commercial tests available that use monoclonal antibodies for the detection of fPSA in
serum. Abbott Diagnostics have 2 commercial assays available (Architect and
AxSYM), with detection limits of 0.3 ng/mL and 0.1 ng/mL, respectively. Diagnostic
Products Corporation have the Immulite assay with detection limits of 0.25 ng/mL.
Roche supply the Elecsys assay with a detection limit of 0.5 ng/mL and Beckman
Coulter assay has limits of 0.2 ng/mL. These commercially available immunoassays
require large expensive equipment at dedicated centres.
Many other novel PSA assay formats have been reported in the literature. (Grubisha et
al., 2003; Soukka et al., 2001; Fernandez-Sanchez et al., 2005; Escamilla-Gomez et al.,
2009). Grubisha et al. (2003) reported sensitivity levels as low as 1 pg/mL using a
sandwich assay on gold nanoparticles coated with a Raman scatterer. Soukka et al.
(2001) reported sensitivity levels in the 0.04 pg/mL range using fluorescently-labelled
nanoparticles. These assays are ultrasensitive and could be employed in detecting levels
of PSA after radical prostatectomy but would not be suitable for early detection of PCa.
Fernandez-Sanchez et al. (2005) reported sensitivities in the region of 1 ng/mL fPSA
using a disposable one-step lateral flow assay but these immunoassays can have
significant drawbacks such as sensitivity. Escamilla-Gomez et al. (2009) described a
method for the detection of fPSA and tPSA using a voltammetric electrochemical
sensor using screen-printed electrodes. The reported sandwich assay format had a range
of detection between 1–10 ng/mL and required a level of expertise in electrochemical
immunoassay development. A major disadvantage when designing electrochemical
immunoassays is the diffusion of the electroactive indicator where the electrodes are in
close proximity resulting in negative results (Escamilla-Gomez et al., 2009). These
reported assays have better sensitivity levels than the sandwich ELISA-based
immunoassay described here, however, there are certain disadvantages associated with
190
using these assays, as previously mentioned. Assay sensitivity could be improved by
performing mutagenesis on the selected antibody fragments (murine and avian). This
would involve site-directed or random mutagenesis improving the affinities and the
assay performance of the antibodies. Assay sensitivity could be further improved by
incorporating the recombinant antibody fragments into an immuno-PCR (Lind et al.,
2005) using a capture antibody with a DNA label for product amplification or
nanoparticle-based immunoassays using a bio-barcode assay for ultrasensitive levels of
detection (Thaxton et al., 2009). These would be suitable immunoassays for detecting
very low levels of PSA after radical prostatectomy where commercial immunoassays
fail to detect PSA levels.
The murine and avian recombinant antibody fragments were analysed on both normal
and tumour prostate cancer cells. BPH is a non-cancerous growth of the prostate gland.
As men get older the prostate grows as a result of proliferation of epithelial cells within
the basal cell layer. BPH is characterized histologically by the presence of discrete
nodules in the periurethral zone of the prostate gland (Edwards, 2008). The
immunohistochemical study showed that both murine and avian recombinant antibody
fragments can be utilised in TMAs for detection of the secreted PSA protein within the
tissue sample. Both antibody fragments bound with less intensity than a polyclonal
positive control but this may be due to the fact that the polyclonal antibody binds many
epitopes on the PSA molecule while the recombinant antibodies have one specific
epitope for binding. This would confirm and support the TMAs results. Varma and coworkers (2002) analysed PCa tissue sections using immunohistochemistry staining and
noticed that polyclonal anti-PSA antibodies were more sensitive than monoclonal
antibodies.
It would also be possible to use a cocktail of recombinant antibodies to different
discrete epitopes on PSA. Expression of recombinant antibody fragments in bacterial
cells offer an unlimited supply of protein at low costs. These are seen as advantageous
characteristics and could substitute polyclonal/monoclonal antibodies for analysing PCa
tissue samples.
191
Chapter 6
Overall Conclusions
192
6.1 Overall conclusions
The aim of the work presented in this thesis was the development of immunoassays for
the detection of PCa. The strategy used involved the generation and characterisation of
recombinant antibody fragments (scFv, scAb and Fab) to fPSA. Individual fragments
were selected by phage display and characterised by ELISA, SDS-PAGE and Western
blotting. Biacore and ELISA-based immunoassays were developed after significant
optimisation for the detection of fPSA.
Chapter 3 describes the methodologies used for the generation of recombinant antibody
fragment libraries. Three antibody libraries were constructed using systems described
by Krebber et al. (1997) and Barbas et al. (2001) from the spleens of immunised mice
and chickens, respectively. Two scFv libraries and one Fab library were constructed and
individual antibody fragments selected by panning. After polyclonal and monoclonal
phage ELISAs confirmed positively-binding antibody fragments individual colonies
were screened by ELISA to determine expression levels of selected antibody fragments.
Comparative analysis between a murine scAb and its scFv counterpart showed
significant differences in expression levels between antibody formats. Avian
recombinant antibody fragments (scFvs and Fabs) were expressed and individual
colonies screened for protein expression by ELISA. The diversity of the original avian
Fab library was clearly observed (Figure 3.21) and colonies selected from round three of
the selection panning process showed that two out of the five analysed antibody
fragments had similar DNA patterns suggesting they could be the same antibody
fragment.
After identification of positive antibody fragments binding to fPSA the next stage
involved additional screening and characterisation of the generated antibody fragments.
A Biacore capture assay was employed to rank 94 expressed scAbs on the basis of their
percentage stability after 5 min dissociation. Eleven antibody fragments with high
percentage stability (above 68%) were selected and subjected to further analysis. The
affinities were calculated using BIAevaluation software and the values were extremely
similar. The most suitable antibody for immunoassay development was selected on a
basis of its competitiveness in an inhibition ELISA. The selected antibody fragment
193
(scAb B5) was expressed in large-scale, purified using IMAC resin and the purity
verified by SDS-PAGE and Western blotting. The avian antibody fragments were
evaluated using the same capture assay approach on the Biacore A100 which allows for
rapid data analysis and interpretation.
Chapter 5 describes the generation of Biacore and ELISA-based immunoassays using
the recombinant antibody fragments selected from Chapter 4. The murine anti-fPSA
scAb was used to develop a subtractive inhibition Biacore assay. The fPSA antigen was
immobilised on the surface of a CM5 sensor chip and the scAb concentration optimised
for assay development using an appropriate flow rate. Decreasing concentrations of
fPSA were incubated with anti-fPSA scAb and passed over the sensor surface. A
decrease in binding was observed with increasing fPSA concentration. Regeneration
studies proved the sensor surface to be stable for over 40 binding cycles with the assay
format demonstrating reproducible and accurate inter-day coefficients of variation
values of less than 20%. Overall, the assay showed good sensitivity and the sensor chip
surface was very stable.
A solid phase inhibition-based ELISA assay was also developed with similar sensitivity
to the Biacore-based assay. The sensitivity was investigated using a sandwich assay
approach to determine whether or not it could be improved. The purified scAb was
biotinylated with the biotin covalently attached to the primary lysine residues of the
antibody. This was captured on the chip using a neutravidin-coated surface and different
concentrations of fPSA captured. The avian anti-fPSA recombinant antibody fragments
(scFv/Fab) were injected over the captured fPSA and binding levels recorded. Suitable
antibody pairs were selected for immunoassay development. A Biacore sandwich assay
proved unsuccessful with no binding observed at low fPSA concentrations (15 ng/mL).
The ELISA-based immunoassay showed a significant improvement in the limit of
detection (1.8 ng/mL). Spiked female serum samples were analysed to investigate any
change in sensitivity using a serum matrix. A slight shift is sensitivity was observed (2.5
ng/mL) but this was expected as the performance of the antibody fragment may be
altered in a complex serum matrix. Samples from individuals presenting with PCa were
analysed by sandwich ELISA and the unknown concentrations determined using a
calibration curve and compared to the values obtained in Beaumont Hospital. The levels
194
of fPSA in each of the samples were in line with reported levels of fPSA. This would
suggest that the generated sandwich immunoassay could potentially be used to detect
fPSA levels in serum. Improvements in assay sensitivity could be achieved using
mutagenesis or employing a novel assay platform as described in Section 5.5. The
immunohistochemistry studies proved that the generated recombinant antibody
fragments (murine scAb and avian scFv) recognised secreted PSA in PCa tissue
samples. This proves that the selected recombinant antibody fragments can be
successfully utilised in TMAs and demonstrates clearly the possibility of using
recombinant antibody fragments to detect markers of disease in tissue samples.
195
Chapter 7
Bibliography
196
Acevedo, B., Perera, Y., Ruiz, M., Rojas, Benetiz, J., Ayala,M. and Gavilondo, J.
(2002). Development and validation of a quantitative ELISA for the measurement of
PSA concentrations. Clin. Chim. Acta, 317, 55-63.
Allard, W. J., Zhou, Z. and Yeung, K. K. (1998). Novel immunoassay for the
measurement of complexed prostate-specific antigen in serum. Clin Chem., 44, 121623.
Anderson, J. (2002). Does PSA testing influence the natural history of prostate cancer?
Eur. Urol. Supp., 1, 3-10.
Andriole, G. L., Levin, D. L. Crawford, E. D., Gelmann, E. P., Pinsky, P. F., Chia, D.,
Kramer, B. S., Reding, D., Church, T. R. Grubb, R. L., Izmirlian, G., Ragard, L. R.,
Clapp, J. D., Prorok, P. C. and Gohagan, J. K. (2005). Prostate cancer screening in the
prostate, lung, colorectal and ovarian (PLCO) cancer screening trial: findings from the
initial screening round of a randomized trial. J. Natl. Cancer Inst., 97, 433-438.
Andris-Widhopf, J., Rader, C., Steinberger, P., Fuller, R. and Barbas, C. F. (2000).
Methods for the generation of chicken monoclonal antibody fragments by phage
display. J. Immunol. Methods, 242, 159-181.
Aus, G., Abbou, C. C., Bolla, M., Heidenreich, A., Schmid, H. P., van Poppel, H.,
Wolff, J. and Zattoni, F. (2005). EAU guidelines on prostate cancer. Eur. Urol., 48,
546-551.
Bartal, A. H., Hirshaut, Y. (1987). Methods for hybridoma formation. The humana
press Inc.
Barbas, C. F., Burton, D. R., Scott, J. K., Silvermann, G. J. (2001). Phage display: A
laboratory manual. Cold spring harbour laboratory press, Cold spring, New York.
197
Besselink, G. A., Kooyman, R. P., van Os, P. J., Engbers, G. H. and Schasfoort, R. B.
(2004). Signal amplification on planar and gel-type sensor surfaces in surface plasmon
resonance-based detection of prostate-specific antigen. Anal Biochem., 333, 165-173.
Birtalan, S., Zhang, Y., Fellouse, F. A., Shao, L., Schaefer, G. and. Sidhu, S. S. (2008).
The intrinsic contributions of tyrosine, serine, glycine and arginine to the affinity and
specificity of antibodies. J. Mol. Biol., 377, 1518-1528.
Bradbury, A. R. M. and Marks, J.D. (2004). Antibodies from phage antibody libraries.
J. Immunol. Methods, 290, 29-49.
Brawer, M. K., Cheli, C. D., Neaman, I. E., Goldblatt, J., Smith, C., Schwartz, M. K.,
Bruzek, D. J., Morris, D. L., Sokoll, L. J., Chan, D. W., Yeung, K. K., Partin, A. W. and
Allard, W. J. (2000). Complexed prostate specific antigen provides significant
enhancement of specificity compared with total prostate specific antigen for detecting
prostate cancer. J. Urol., 163, 1476-1480.
Breul, J., Pickl, U. and Hartung, R. (1994). Prostate-specific antigen in urine. Eur.
Urol., 26, 18-21.
Bruckler, D. R., Park, A., Viswanathan, M., Hoet, R. M. and Ladner, R. C. (2008).
Screening isolates from antibody phage-display libraries. Drug Discovery Today, 13,
318-323.
Butch, A. W., Crary, D. and Yee, M. (2001). Analytical performance of the Roche toal
and free PSA assays on the Elecsys 2010 immunoanalyzer. Clin. Biochem., 35, 143145.
Cao, C., Kim, J. P., Kim, B. W., Chae, H., Yoon, H. C., Yang, S. S. and Sim, S. J.
(2006). A strategy for sensitivity and specificity enhancements in prostate specific
antigen-alpha (1)-antichymotrypsin detection based on surface plasmon resonance.
Biosens. Bioelectron., 21, 2106-2113.
198
Canziani, G. A., Klakamp, S. and Myszka, D. G. (2004). Kinetic screening of
antibodies from crude hybridoma samples using Biacore. Anal. Biochem., 325, 301-307.
Carvalho A. L., Sanz, L., Barettino, D., Romero, A., Calvete, J. J., and Roma, M. J.
(2002). Crystal structure of a prostate kallikrein isolated from stallion seminal plasma:
A homologue of human PSA. J. Mol. Biol., 322, 325-337.
Casper, D., Bukhtiyarova, M. and Springman, E. B. (2004). Comparison of p38 kinase
inhibitor binding affinity and kinetics using Biacore. Anal. Biochem., 325, 126-136.
Catalona, W. J., Bartsch, G., Rittenhouse, H. G., Evans, C. L., Linton, H. J.,
Amirkhan, A., Horninger, W., Klocker, H. and Mikolajczyk, S. D. (2003). Serum pro
prostate specific antigen improves cancer detection compared to free and complexed
prostate specific antigen in men with prostate specific antigen 2 to 4 ng/ml. J. Urol.,
170, 2181-2195.
Catalona, W. J., Partin, A. W., Slawin, K. M., Brawer, M. K., Flanigan, R. C., Patel,
A., Richie, J. P., deKernion, J. B., Walsh, P. C., Scardino, P. T., Lange, P. H., Subong,
E. N., Parson, R. E., Gasior, G. H., Loveland, K. G. and Southwick, P. C. (1998). Use of
the percentage of free prostate-specific antigen to enhance differentiation of prostate
cancer from benign prostatic disease: a prospective multicenter clinical trial. JAMA.,
279, 1542-1547.
Catalona, W. J., Smith, D. S., Wolfert, R. L., Wang, T. J., Rittenhouse, H. G., Ratliff,
T. L. and Nadler, R. B. (1995). Evaluation of percentage of free serum prostate-specific
antigen to improve specificity of prostate cancer screening. JAMA ., 274, 1214-20.
Catalona, W. J., Richie, J. P., Ahmann, F. R., Hudson, M. A., Scardino, P. T.,
Flanigan, R. C., deKernion, J. B., Ratliff, T. L., Kavoussi, L. R., Dalkin, B. L., et al.
(1994). Comparison of digital rectal examination and serum prostate specific antigen in
the early detection of prostate cancer: results of a multicenter clinical trial of 6,630 men.
J. Urol., 151, 1283-1290.
199
Catalona, W. J., Smith, D. S., Ratliff, T.L., Dodds, K. M., Coplen, D. E., Yuan, J. J.,
Petros, J. A. and Andriole, G. L. (1991). Measurement of prostate-specific antigen in
serum as a screening test for prostate cancer. N. Engl. J. Med., 324, 1156-1161.
Charnow, J. (2008). Statins raise PCa odds in obese men. Amer. J. Epidemiol., 168,
250-260.
Chen, Z., Chen, H., and Stamey, T. A. (1997). Prostate-specific antigen in benign
prostatic hyperplasia: purification and characterization. J. Urol. 157, 2166-2170.
Chen, L., Stacewicz-Sapuntzakis, M., Duncan, C., Sharifi, R., Ghosh, L., van Breemen,
R., Ashton, D. and Bowen, P. E. (2001). Oxidative DNA damage in prostate cancer
patients consuming tomato sauce-based entrees as a whole-food intervention. J. Natl.
Cancer. Inst., 93, 1872-1879.
Cherian, S., Gupta, R. K., Mullin, B. C. and Thundat, T. (2003). Detection of heavy
metal ions using protein-functionalized microcantilever sensors. Biosens. Bioelectron.,
19, 411-416.
Cuong, C., Kim, J. P., Kim, B. W., Chae, H. C., Yang, S. S. and Sang, S. J. (2005). A
strategy for sensitivity and specificity enhancements in prostate specific antigen-α1antichymotrypsin detection based on surface plasmon resonance. Biosens. Bioelectron.,
21, 2106-2113.
Darson, M. F., Pacelli, A., Roche, P., Rittenhouse, H. G., Wolfert, R. L., Saeid, M. S.,
Young, C. Y., Klee, G. G., Tindall, D. J., and Bostwick, D. G. (1999). Human glandular
kallikrein 2 expression in prostate adenocarcinoma and lymph node metastases. Urol.,
53, 939-944.
Delves, P. J., Lund, T. and Roitt, I. M. (1997). Can epitope-focused vaccines select
advantageous immune responses? Mol. Med. Today., 3, 55-60.
200
Denmeade, S. R. and Isaacs, J. T. (2004). Development of prostate cancer treatment:
the good news. Prostate, 58, 211-224.
Dennis, L. K., Lynch, C. F. and Torner, J.C. (2002). Epidemiologic association
between prostatitis and prostate cancer. Urol., 60, 78-83.
Diamandis, E. P., Okui, A., Mitsui, S., Luo, L. Y., Soosaipillai, A., Grass, L.,
Nakamura, T., Howarth, D. J. and Yamaguchi, N. (2002). Human Kallikrein 11: A new
biomarker of prostate and ovarian carcinoma. Cancer Res., 62, 295-300.
Diamandis, E. P. and Yousef, G. M. (2002). Human tissue kallikreins: a family of new
cancer biomarkers. Clin. Chem., 48, 1198-1205.
Dillon, P. P., Manning, B. M., Daly, S. J., Killard, A. J. and O'Kennedy, R. (2003).
Production of a recombinant anti-morphine-3-glucuronide single-chain variable
fragment (scFv) antibody for the development of a "real-time" biosensor-based
immunoassay. J. Immunol. Methods, 276, 151-61.
Djavan, B., Fong, Y. K., Remzi, M., Fakhari, M. and Marberger, M. (2004). New
serum and urinary markers for prostate cancer detection in the new millennium. Eur.
Urol. Supp., 3, 25-32.
Don, R. H., Cox, P. T., Wainwright, B. J., Baker, K. and Mattick, J. S. (1991).
'Touchdown' PCR to circumvent spurious priming during gene amplification. Nucleic
Acids Res., 19, 4008.
Drukier, A. K., Ossetrova, N., Schors, E., Brown, L. R., Tomaszewski, J., Sainsbury, R
and Godovac-Zimmermann, J. (2005). Ultra-sensitive immunoassays using multiphoton-detection in diagnostic proteomics of blood. J. Proteome Res., 4, 2375-2378.
Edwards, J.L. (2008). Diagnosis and management of benign prostatic hyperplasia.
Amer. Family Phys., 77, 1403-1410.
201
Escamilla-Gomez, V., Hernandez-Santos, D., Gonzalez-Garcia, M. B., PingarronCarrazon, J. M. and Costa-Garcia, A. (2009). Simultaneous detection of free and total
prostate specific antigen on a screen-printed electrochemical dual sensor. Biosens.
Bioelectron., 24, 2678-2683.
Fernandez-Sanchez, C., McNeil, C. J., Rawson, K., Nilsson, O., Leung, H. Y. and
Gnanapragasam, V. (2005). One-step immunostrip test for the simultaneous detection of
free and total prostate specific antigen in serum. J. Immunol. Methods, 307, 1-12.
Finlay, W. J. J., Shaw, I., Reilly, J. P. and Kane, M. (2006). Generation of high-affinity
chicken single-chain fv antibody fragments for measurement of the Pseudonitzschia
pungens toxin Domoic acid. Appl. Environ. Microbiol. 72, 3343-49.
Finne, P., Zhang, W. M., Auvinen, A., Leinonen, J., Maattanen, L., Rannikko, S.,
Tammela, T. L. and Stenman, U. H. (2000). Use for the complex between prostate
specific antigen and alpha-1-protease inhibitor for screening prostate cancer. J. Urol.
164, 1956–1960.
Frederix, F., Friedt, J. M., Choi, K. H., Laureyn, W., Campitelli, A., Mondelaers, D.,
Maes, G. and Borghs, G. (2003). Biosensing based on light absorption of nanoscaled
gold and silver particles. Anal. Chem., 75, 6894-6900.
Gann, P. and Giovannucci, E. (2005) Nutrition, exercise and prostate cancer
http://www.prostatecancerfoundation.org/atf/cf/%7B705B3273-F2EF-4EF6-A653E15C5D8BB6B1%7D/Nutrition_Guide.pdf.
Gann, P. H., Ma, J., Giovannucci, E., Willett, W., Sacks, F. M., Hennekens, C. H. and
Stampfer, M. J. (1999). Lower prostate cancer risk in men with elevated plasma
lycopene levels: results of a prospective analysis. Cancer Res., 59, 1225-30.
Garattini, S. and Shore, P. A. (1966). Advances in pharmacology. Academic Press Inc.,
New York.
202
Giovannucci, E., Rimm, E. B., Colditz, G. A., Stampfer, M. J., Ascherio, A., Chute, C.
C. and Willett, W. C. (1993). A prospective study of dietary fat and risk of prostate
cancer. J. Natl. Cancer Inst ., 85, 1538-1540.
Gram, H., Marconi, L. A., Barbas, C. F., Collet, T. A., Lerner, R. A. and Kang, A. S.
(1992). In vitro selection and affinity maturation of antibodies from a naive
combinatorial immunoglobulin library. Proc. Natl. Acad. Sci. USA, 89, 3576-3580.
Graves, H. C., Sensabaugh, G. F. and Blake, E. T. (1985). Postcoital detection of a
male-specific semen protein. Application for the investigation of rape. N. Engl. J. Med.,
312, 338-343.
Gregorakis, A. K., Malovrouvas, D., Stefanakis, S., Petrakic, K., and Scorilas, A.
(2005). Free/Total PSA (F/T ratio) kinetics in patients with clinically localized prostate
cancer undergoing radical prostatectomy. Eur. Urol. Supp., 5, 472-478.
Greunke, K., Spillner, E., Braren, I., Seismann, H., Kainz, S., Hahna, U., Grunwald, T.
and Bredehorst, R. (2006).Bivalent monoclonal IgY antibody formats by conversion of
recombinant antibody fragments. J. Biotech., 124, 446–456.
Grubisha, D. S., Lipert, R. J., Park, H. Y., Driskell, J. and Porter, M. D. (2003).
Femtomolar detection of prostate-specific antigen: an immunoassay based on surfaceenhanced Raman scattering and immunogold labels. Anal. Chem., 75, 5936-5943.
Haese, A., Graefen, M., Steuber, T., Becker, C., Pettersson, K., Piironen, T., Noldus, J.,
Huland, H. and Lilja, H. (2001). Human glandular kallikrein 2 levels in serum for
discrimination of pathologically organ-confined from locally-advanced prostate cancer
in total PSA-levels below 10 ng/ml. Prostate, 49, 101-109.
Hayhurst, A. (2000). Improved expression characteristics of single-chain Fv fragments
when fused downstream of the Escherichia coli maltose-binding protein or upstream of
a single immunoglobulin-constant domain. Protein Exp. Purif. 18, 1-10.
203
Hayhurst, A., Happe, S., Mabry, R., Koch, Z., Iverson, B.L. and Georgiou, G. (2003).
Isolation and expression of recombinant antibody fragments to the biological warfare
pathogen Brucella melitensis. Immunol. Methods, 276: 185-196.
Hayward, S. W. and Cunha, G. R. (2000). The prostate: development and physiology.
Radiol. Clin. North Am., 38, 1-14.
Healy, D. A., Hayes, C. J., Leonard, P., McKenna, L. and O‘Kennedy, R. (2007).
Biosensor developments: application to prostate-specific antigen detection. Trends in
Biotech., 3, 125-131.
Hernandez, V. Q., Juárez-González, V. R., Ortíz-León, M., Sánchez, R., Possani, L. D.
and Becerril, B. (2006). The change of the scFv into the Fab format improves the
stability and in vivo toxin neutralization capacity of recombinant antibodies. Mol.
Immunol., 44, 1307-1315.
Hessels, D., Verhaegh, G. W., Schalken, J. A. and Witjes, J. A. (2004). Applicability of
biomarkers in the early diagnosis of prostate cancer. Expert Rev. Mol. Diagn., 4, 513526.
Hilz, H., Noldus, J., Hammerer, P., Buck, F., Luck, M. and Huland, H. (1999).
Molecular heterogeneity of free PSA in sera of patients with benign and malignant
prostate tumors. Eur. Urol., 36, 286-292.
Hochmeister, M. N., Borer, U., Budowle, B., Dirnhofer, R., Gehrig, C., Rudin, O., and
Thali, M. (1999). Evaluation of prostate-specific antigen (PSA) membrane test assays
for the forensic identification of seminal fluid, J. Forensic Sci., 44, 1057-1060.
Hoffman, R.M., Gilliland, F.D., Eley, J.W., Harlan, L.C., Stephenson, R.A., Stanford,
J.L., Albertson, P.C., Hamilton, A.S., Hunt, W.C. and Potosky, A.L. (2001). Racial and
ethnic differences in advanced-stage prostate cancer: the Prostate Cancer Outcomes
Study. J. Natl. Cancer Inst., 93, 388-395.
204
Hoogenboom, H. R., Henderikx, P. and de Haard, H. (1998). Creating and engineering
human antibodies for immunotherapy. Adv. Drug Deliv. Rev., 31, 5-31.
http://www.mamashealth.com/organs/prostate.asp.
http://www.psa-rising.com/prostatecancer/prostate.htm
Hsing, A. W. and Chokkalingam, A. P. (2006). Prostate cancer epidemiology. Frontiers
Biosci., 11, 1388-1413.
Hsing, A. W., Tsao, L. and Devesa, S. S. (2000). International trends and patterns of
prostate cancer incidence and mortality. Int. J. Cancer, 85, 60-67.
Huang, L., Reekmans, G., Saerens, D., Friedt, J. M., Frederix, F., Francis, L.,
Muyldermans, S., Campitelli, A. and Van Hoof, C. (2005). Prostate-specific antigen
immunosensing based on mixed self-assembled monolayers, camel antibodies and
colloidal gold enhanced sandwich assays. Biosens. Bioelectron., 21, 483-490.
Huhtinen, P., Soukka, T., Lovgren, T. and Harma, H. (2004). Immunoassay of total
prostate-specific antigen using europium(III) nanoparticle labels and streptavidin-biotin
technology. J. Immunol. Methods, 294, 111-122.
Hutchinson, L. M., Chang, E. L., Becker, C. M., Shih, M. C., Brice, M., DeWolf, W.
C., Gaston, S. M. and Zetter, B. R. (2005a). Use of thymosin beta15 as a urinary
biomarker in human prostate cancer. Prostate, 64, 116-127.
Hutchinson, L. M., Chang, E. L., Becker, C. M., Ushiyama, N., Behonick, D., Shih, M.
C., DeWolf, W. C., Gaston, S. M. and Zetter, B. R. (2005b). Development of a sensitive
and specific enzyme-linked immunosorbent assay for thymosin beta15, a urinary
biomarker of human prostate cancer. Clin. Biochem., 38, 558-571.
Hwang, K. S., Lee, J. H., Park, J., Yoon, D. S., Park, J. H. and Kim, T. S. (2004). Insitu quantitative analysis of a prostate-specific antigen (PSA) using a nanomechanical
PZT cantilever. Lab Chip, 4, 547-552.
205
Ippolito, G. C., Schelonka, R. L., Zemlin, M., Ivanov, I. I., Kobayashi, R., Zemlin,
Gartland, G. L., Nitschke, L., Pelkonen, J., Fujihashi, K., Rajewsky, K. and Schroeder
Jr. H. W. (2006). Forced usage of positively charged amino acids in immunoglobulin
CDR-H3 impairs B cell development and antibody production. J. Experim. Med., 6,
1567-1578.
Irving, R. A., Korett, A. A. and Hudson, P. J. (1996). Affinity maturation of
recombinant antibodies using E. coli mutator cells. Immunotechnol., 2, 127-143.
Irani, J., Salomon, L., Soulie, M., Zlotta, A., de la Taille, A., Dore, B. and Millet, C.
(2005). Urinary/serum prostate-specific antigen ratio: comparison with free/total serum
prostate-specific antigen ratio in improving prostate cancer detection. Urol., 65, 533537.
Ismail, H. A., Pollak, M., Behlouli, H., Tanguay, S., Begin, L. R. and Aprikian, A. G.
(2003). Serum insulin-like growth factor (IGF)-1 and IGF-binding protein-3 do not
correlate with Gleason score or quantity of prostate cancer in biopsy samples. BJU Int.,
92, 699-702.
Jacobs, E. J., Rodriguez, C., Mondul, A. M., Connell, C. J., Henley, S. J., Calle, E. E.
and Thun, M. J. (2005). A large cohort study of aspirin and other nonsteroidal antiinflammatory drugs and prostate cancer incidence. J. Natl. Cancer Inst., 97, 975-980.
Janeway, C., Travers, P., Walport, M., Capra, J. D. (1999). In immunology: the
immune system in health and disease. 4th edition New York, Garland Publishing.
Johnson, E. D. and Kotowski, T. M. (1993). Detection of prostate-specific antigen by
ELISA. J. Forensic Sci., 38, 250-258.
Jimenez-Verdejo, A., Osuna, E., Garcia-Olivares, E. and Luna, A. (1994). Study of the
enzymatic activity of GGT, LDH, PAP and PSA in semen stains: application to age
calculation. Forensic Sci. Int., 68, 7-15.
206
Jung, K., Brux, B., Lein, M., Rudolph, B., Kristiansen, G., Hauptmann, S., Schnorr, D.,
Loening, S. A. and Sinha, P. (2000a). Molecular forms of prostate-specific antigen in
malignant and benign prostatic tissue: biochemical and diagnostic implications. Clin.
Chem., 46, 47-54.
Jung, K., Elgeti, U., Lein, M., Brux, B., Sinha, P., Rudolph, B., Hauptmann, S.,
Schnorr, D., Loening, S. A. (2000b). Ratio of free or complexed prostate-specific
antigen (PSA) to total PSA: which ratio improves differentiation between benign
prostatic hyperplasia and prostate cancer? Clin. Chem., 46, 55-62.
Katsamba, P. S., Navratilova, I., Calderon-Carcia, M., Fan, L., Thornton, K., Zhu, M.,
Vandel Bos, T., Forte, C., Friend, D., Laird-Offringa, I., Tavares, G., Whatley, J., Shi,
E., Widom, A., Lindquist, K. S., Klakamp, S., Drake, A., Bohmann, D., Roell, M.,
Rose, L., Dorocke, J., Roth, B., Luginbuhl, B. and Myszka, D. G. (2006). Kinetic
analysis of a high-affinity antibody/antigen interaction performed by multiple Biacore
users. Anal. Biochem., 352, 208-221.
Khaldi, N., Miras, A., Botti, K., Benali, L. and Gromb, S. (2004). Evaluation of three
rapid detection methods for the forensic identification of seminal fluid in rape cases. J.
Forensic Sci., 49, 749-753.
Khan, M. A., Partin, A. W., Rittenhouse, H. G., Mikolajczyk, S. D., Sokoll, L. J., Chan,
D. W. and Veltri, R. W. (2003). Evaluation of pro-PSA for the early detection of
prostate cancer in men with a total PSA range of 4.0-10 ng/ml. J. Urol., 170, 723-726.
Klassen, A. C. and Platz, E. A. (2006).What can geography tell us about prostate
cancer? Am. J. Prev. Med., 30, S7-15.
Klotz, L. and Teahan, S. (2006). Current role of PSA kinetics in the management of
patients with prostate cancer. Eur. Urol. Supp., 5, 472-478.
Kohler, G. and Milstein, C. (1975). Continuous cultures of fused cells secreting
antibody of predefined specificity. Nature (London), 256, 495-497.
207
Koide, S. and Sidhu, S. S. (2009). The importance of being tyrosine: lessons in
molecular recognition from minimalist synthetic binding proteins. ACS Chem. Biol., 4,
325-334.
Khosravi, M. J., Papanastasiou-Diamandi, A. and Mistry, J. (1995). An ultrasensitive
immunoassay for prostate-specific antigen based on conventional colorimetric
detection. Clin. Biochem., 28, 407-414.
Krebber, A., Bornhauser, S., Burmester, J., Honegger, A., Willuda, J., Bosshard, H. R.
and Pluckthun, A. (1997). Reliable cloning of functional antibody variable domains
from hybridomas and spleen cell repertoires employing a reengineered phage display
system. J. Immunol. Methods, 201, 35-55.
Krumholtz, J. S., Carvalhal, G. F., Ramos, C. G., Smith, D. S., Thorson, P., Yan, Y.,
Humphrey, P. A., Roehl, K. A. and Catalona, W. J. (2002). Prostate-specific antigen
cutoff of 2.6 ng/mL for prostate cancer screening is associated with favorable
pathologic tumor features. Urol., 60, 469-473.
Kumar, A., Mikolajczyk, S. D., Goel, A. S., Millar, L. S. and Saedi, M. S. (1997).
Expression of pro form of prostate-specific antigen by mammalian cells and its
conversion to mature, active form by human kallikrein 2. Cancer Res., 57, 3111-3114.
Kuriyama, M., Wang, M. C., Papsidero, L. D., Killian, C. S., Shimano, T., Valenzuela,
L., Nishiura, T., Murphy, G. P. and Chu, T. M. (1980). Quantitation of prostate-specific
antigen in serum by a sensitive enzyme immunoassay. Cancer Res., 40, 4658-4662.
Kwiatkowski, M. K., Recker, F., Piironen, T., Pettersson, K., Otto, T., Wernli, M. and
Tscholl, R. (1998). In prostatism patients the ratio of human glandular kallikrein to free
PSA improves the discrimination between prostate cancer and benign hyperplasia
within the diagnostic 'grey zone' of total PSA 4 to 10 ng/ml. Urol., 52, 360-365.
Labrie, F., Candas, B., Dupont, A., Cusan, L., Gomez, J. L., Suburu, R. E., Diamond,
P., Levesque, J. And Belanger, A. (1999). Screening decreases prostate cancer death:
208
first analysis of the 1988 Quebec prospective randomized controlled trial. Prostate, 38,
83-91.
Laux, D. L. and Custis, S. E. (2003). Forensic Detection of Semen III. Detection of
PSA Using Membrane Based Tests: Sensitivity Issues with Regards to the Presence of
PSA in Other Body Fluids. Submitted MAFS newsletter.
Lee, J. H., Hwang, K. S., Park, J., Yoon, K. H., Yoon, D. S. and Kim, T. S. (2005).
Immunoassay of prostate-specific antigen (PSA) using resonant frequency shift of
piezoelectric nanomechanical microcantilever. Biosens. Bioelectron., 20, 2157-2162.#
Leinonen, J., Niemelä, P., Lövgren, J., Bocchi, L. Pettersson, K., Nevanlinna, H. and
Stenman, U. H. (2004). Characterization of monoclonal antibodies against prostate
specific antigen produced by genetic immunization. J. Immunol. Methods, 289, 157167.
Levesque, M., Hu, H., D'Costa, M. and Diamandis, E. P. (1995). Prostate-specific
antigen expression by various tumors. J. Clin. Lab. Anal., 9, 123-128.
Leonard, P., Safsten, P., Hearty, S., McDonnell, B., Finlay, W. and O‘Kennedy, R.
(2007). High throughput ranking of recombinant avian scFv antibody fragments from
crude lysates using Biacore A100. J. Immunol. Methods, 323, 172-179.
Lichtenstein, P., Holm, N. V., Verkasalo, P. K., Iliadou, A., Kaprio, J., Koskenvuo, M.,
Pukkala, E., Skytthe, A. and Hemminki, K. (2000). Environmental and heritable factors
in the causation of cancer--analyses of cohorts of twins from Sweden, Denmark, and
Finland. N. Engl. J. Med., 343, 78-85.
Lilja, H., Ulmert, D. and Vickers, A. J. (2008). Prostate-specific antigen and prostate
cancer: prediction, detection and monitoring. Nature Reviews Cancer, 8, 268-278.
209
Lilja, H., Christensson, A., Dahlen, U., Matikainen, M. T., Nilsson, O., Pettersson, K.
and Lovgren, T. (1991). Prostate-specific antigen in serum occurs predominantly in
complex with alpha 1-antichymotrypsin. Clin. Chem., 37, 1618-1625.
Lin, H., and Cornish, V. W. (2002). Screening and selection methods for large-scale
analysis of protein function. Angew. Chem, Int. Ed., 41, 4402-4425.
Lind, K. and Kubista, M. (2005). Development and evaluation of three real-time
immuno-PCR assemblages for quantification of PSA. J. Immunol. Methods, 304, 107116.
Linton, H. J., Marks, L. S., Millar, L. S., Knott, C. L., Rittenhouse, H. G. and
Mikolajczyk, S. D. (2003). Benign prostate-specific antigen (BPSA) in serum is
increased in benign prostate disease. Clin. Chem., 49, 253-259.
Little M., Kipriyanov S. M., Le Gall F. and Moldenhauer G. (2000). Of mice and men:
hybridoma and recombinant antibodies. Immunol. Today, 21, 364-370.
Low, N. M., Holliger, P. H. and Winter G. (1996). Mimicking somatic hypermutation:
Affinity maturation of antibodies displayed on bacteriophage using a bacterial mutator
strain. J. Mol. Biol., 260, 359-368.
Lutz, G. L., Zuber, E., Jean Witz, J. and Regenmortel, M. H. V. (1997).
Thermodynamic analysis of antigen–antibody binding using biosensor measurements at
different temperatures. Anal. Biochem., 246, 123-132.
Lytton, B. (2001). Prostate cancer: a brief history and the discovery of hormonal
ablation treatment. J. Urol., 165, 1859-1862.
Magee, J. A., Araki, T., Patil, S., Ehrig, T., True, L., Humphrey, P. A., Catalona, W. J.,
Watson, M. A. and Milbrandt, J. (2001). Expression profiling reveals hepsin
overexpression in prostate cancer. Cancer Res., 61, 5692-5696.
210
Mahabeer, R. and Bhoola, K. D. (2000). Kallikrein and kinin receptor genes.
Pharmacol. Ther., 88, 77-89.
Maher, J., Vintiner, S., Elliot, D. and Melia, L. (2002). Evaluation of the BioSign PSA
membrane test for the identification of semen stains in forensic casework. N. Z. Med. J.,
115, 48-49.
Marks, J. D., Griffiths, A. D., Malmqvist, M., Clackson, T. P., Bye, J. M. and Winter,
G. (1992). By-passing immunisation: Building high affinity human antibodies by chain
shuffling. Nature Biotech., 10, 779-783.
Matsumoto, K., Konishi, N., Hiasa, Y., Kimura, E., Takahashi, Y., Shinohara, K. and
Samori, T. (1999). A highly sensitive enzyme-linked immunoassay for serum free
prostate-specific antigen (f-PSA). Clin. Chim. Acta ., 281, 57-69.
Matsumoto, K., Konishi, N., Samori, T., Kimura, E., Doi, M., Kato, S. and Yuki, Y.
(2000). ELISA for a complexed antigen with a monoclonal antibody blocking reaction
with the free antigen—assay-specific for complexed prostate-specific antigen. J.
Immunol. Methods, 234, 99-106.
McCafferty J, Griffiths, A. D, Winter, G. and Chiswell D. J. (1990). Phage antibodies:
filamentous phage displaying antibody variable domains. Nature, 348, 552–554.
Mikolajczyk, S. D., Song, Y., Wong, J. R., Matson, R. S. and Rittenhouse, H. G.
(2004). Are multiple markers the future of prostate cancer diagnostics? Clin. Biochem.,
37, 519-528.
Mikolajczyk, S. D. and Rittenhouse, H. G. (2003). Pro PSA: a more cancer-specific
form of prostate specific antigen for the early detection of prostate cancer. Keio. J.
Med., 52, 86-91.
Mikolajczyk, S. D., Marker, K. M., Millar, L. S., Kumar, A., Saedi, M. S., Payne, J. K.,
Evans, C. L., Gasior, C. L., Linton, H. J., Carpenter, P. and Rittenhouse, H. G. (2001).
211
A truncated precursor form of prostate-specific antigen is a more specific serum marker
of prostate cancer. Cancer Res., 61, 6958-6963.
Mikolajczyk, S. D., Millar, L. S., Wang, T. J., Rittenhouse, H. G., Wolfert, R. L.,
Marks, L. S., Song, W., Wheeler, T. M. and Slawin, K. M. (2000). "BPSA," a specific
molecular form of free prostate-specific antigen, is found predominantly in the
transition zone of patients with nodular benign prostatic hyperplasia. Urol., 55, 41-45.
Mikolajczyk, S. D., Grauer, L. S., Millar, L. S., Hill, T. M., Kumar, A., Rittenhouse, H.
G., Wolfert, R. L. and Saedi, M. S. (1997). A precursor form of PSA (pPSA) is a
component of the free PSA in prostate cancer serum. Urol., 50, 710-714.
Miller, D. C., Hafez, K. S., Stewart, A., Montie, J. E. and Wei, J. T. (2003). Prostate
carcinoma presentation, diagnosis, and staging: an update form the National Cancer
Data Base. Cancer, 98, 1169-1178.
Mistry, K., and Cable, G. (2003). Meta-analysis of prostate-specific antigen and Digital
rectal examination as screening tests for prostate carcinoma. J. Am. Board Family
Pract., 16, 95-101.
Mitchell, I. D., Croal, B. L., Dickie, A., Cohen, N. P. and Ross, I.
(2001). A
prospective study to evaluate the role of complexed prostate specific antigen and
free/total prostate specific antigen ratio for the diagnosis of prostate cancer. J. Urol.,
165, 1549-1553.
Murphy, M., Jason-Moller, L. and Bruno, J. (2006). Using Biacore to measure the
binding kinetics of an antibody-antigen interaction. Current Protocols in Protein
Science, Chapter 19. 19.14.1-19.4.17 (John Wiley & Sons).
Nadler, R. B., Humphrey, P. A., Smith, D. S., Catalona, W. J. and Ratliff, T. L.
(1995). Effect of inflammation and benign prostatic hyperplasia on elevated serum
prostate specific antigen levels. J. Urol., 154, 407-413.
212
Nakamura, T., Scorilas, A., Stephan, C., Jung, K., Soosaipillai, A. R. and Diamandis,
E. P. (2003). The usefulness of serum human kallikrein11 (hK11) for discriminating
between prostate cancer and benign prostatic hyperplasia. Cancer Res., 63, 6543-6546.
Naya, Y., Fritsche, H. A., Bhadkamkar, V. A., Mikolajczyk, S. D., Rittenhouse, H. G.
and Babaian, R. J. (2005). Evaluation of precursor prostate-specific antigen isoform
ratios in the detection of prostate cancer. Urol. Oncol., 23, 16-21.
Ni, J., Lipert, R. J., Dawson, G. B. and Porter, M. D. (1999). Immunoassay readout
method using extrinsic Raman labels adsorbed on immunogold colloids. Anal. Chem.,
71, 4903-4908.
Noldus, J., Chen, Z. and Stamey, T. A. (1997). Isolation and characterization of free
form prostate specific antigen (f-PSA) in sera of men with prostate cancer. J. Urol.,
158, 1606-1609.
Nurmikko, P., Pettersson, K., Piironen, T., Hugosson, J. and Lilja, H. (2001).
Discrimination of prostate cancer from benign disease by plasma measurement of intact,
free prostate-specific antigen lacking an internal cleavage site at Lys145-Lys146. Clin.
Chem., 47, 1415-1423.
Nurmikko, P., Vaisanen, V., Piironen, T., Lindgren, S., Lilja, H. and Pettersson, K.
(2000). Production and characterization of novel anti-prostate-specific antigen (PSA)
monoclonal antibodies that do not detect internally cleaved Lys145-Lys146 inactive
PSA. Clin. Chem., 46, 1610-1618.
Oberpenning, F., Weining, C., Brandt, B., De Angelis, G., Heinecke, A., Hamm, M.,
Stieber, P., Hertle, L., Schmid, H.P. and Semjonow, A. (2002). Combining free and
total prostate specific antigen assays from different manufacturers: the pitfalls. Eur.
Urol., 42, 577-82.
Obiezu, C. V. and Diamandis, E. P. (2005). Human tissue kallikrein gene family:
applications in cancer. Cancer Lett., 224, 1-22.
213
Obiezu, C. V., Soosaipillai, A., Jung, K., Stephan, C., Scorilas, A., Howarth, D. H. and
Diamandis, E. P. (2002). Detection of human kallikrein 4 in healthy and cancerous
prostatic tissues by immunofluorometry and immunohistochemistry. Clin. Chem., 48,
1232-1240.
Oh, M. Y., Joo, H. Y., Hur, B. U., Jeong, Y. H. and Cha, S. H. (2007). Enhancing
phage display of antibody fragments using gIII-amber suppression. Gene, 386, 81-89.
Özena, H., and Sözen, S. (2006). PSA isoforms in prostate cancer detection. Eur. Urol.
Supp., 5, 495-499.
Pakioures M., Borgono, C. and Diamandis, E. (2007). Human tissue kallikreins: The
cancer biomarker family. Cancer letters, 249, 61-79.
Papotti, M., Paties, C., Peveri, V., Moscuzza, L. and Bussolati, G. (1989).
Immunocytochemical detection of prostate-specific antigen in skin adnexal and breast
tissues and tumors. Basic Appl. Histochem., 33, 25-29.
Partin, A. W., Catalona, W. J., Finlay, J. A., Darte, C., Tindall, D. J., Young, C. Y.,
Klee, G. G., Chan, D. W., Rittenhouse, H. G., Wolfert, R. L. and Woodrum, D. L.
(1999). Use of human glandular kallikrein 2 for the detection of prostate cancer:
preliminary analysis. Urol., 54, 839-845.
Peter, J., Unverzagt, C., Krogh, T. N., Vorm, O. and Hoesel, W. (2001). Identification
of precursor forms of free prostate-specific antigen in serum of prostate cancer patients
by immunosorption and mass spectrometry. Cancer Res., 61, 957-962.
Peyromaure, M., Fulla, Y., Debre, B. and Dinh-Xuan, A. T. (2005). Pro PSA: a "pro
cancer" form of PSA? Med. Hypotheses, 64, 92-95.
Popkov, M., Rader, C. and Barbas, III, C. F. (2004). Isolation of human prostate cancer
cell reactive antibodies using phage display technology. J. Immunol. Methods, 291, 137151.
214
Roddam, A. W., Duffy, M. J., Hamdy, F. C., Ward, A. M., Patnick, J., Price, C. P.,
Rimmer, J., Sturgeon, C., White, P. and Allen, N. E. (2005). Use of prostate-specific
antigen (PSA) isoforms for the detection of prostate cancer in men with a PSA level of
2-10 ng/ml: systematic review and meta-analysis. Eur. Urol., 48, 386-99.
Rella, R., Spadavecchiaa, J., Manera, M. G., Siciliano, P., Santino, A., and Mita, G.
(2004). Liquid phase SPR imaging experiments for biosensors applications. Biosens.
Bioelectron., 20, 1140-1148.
Rhodes, D. R., Sanda, M. G., Otte, A. P., Chinnaiyan, A. M. and Rubin, M. A. (2003).
Multiplex biomarker approach for determining risk of prostate-specific antigen-defined
recurrence of prostate cancer. J. Natl. Cancer Inst., 95, 661-668.
Roitt, I. M. (2006). Roitt‘s essential immunology. 11th edition. Publishers, Maden mass,
Oxford, Blackwell.
Rojas, G., Almagro, J. C., Acevedo, B. And Gavilondo, J. V. (2002). Phage antibody
fragments library combining a single human light chain variable region with immune
mouse heavy chain variable regions. J. Biotechnol., 94, 287-298.
Rubin, M. A., Mucci, N. R., Figurski, J., Fecko, A., Pienta, K. J. and Day, M. L.
(2001). E-cadherin expression in prostate cancer: a broad survey using high-density
tissue microarray technology. Hum. Pathol., 32, 690-697.
Rubin, M. A., Zhou, M., Dhanasekaran, S. M., Varambally, S., Barrette, T. R., Sanda,
M. G., Pienta, K. J., Ghosh, D., and Chinnaiyan, A. M. (2002). Alpha-methylacyl
coenzyme A racemase as a tissue biomarker for prostate cancer. JAMA., 287, 16621670.
Saerens, D., Frederix, F., Reekmans, G., Conrath, K., Jans, K., Brys, L., Huang, L.,
Bosmans, E., Maes, G., Borghs, G. and Muyldermans, S. (2005). Engineering camel
single-domain antibodies and immobilization chemistry for human prostate-specific
antigen sensing. Anal. Chem., 77, 7547-7555.
215
Sano, T., Smith, C. L. and Cantor, C. R. (1992). Immuno-PCR: very sensitive antigen
detection by means of specific antibody-DNA conjugates. Science, 258, 120-122.
Sato, I., Sagi, M., Ishiwari, A., Nishijima, H., Ito, E. and Mukai, T. (2002). Use of the
"SMITEST" PSA card to identify the presence of prostate-specific antigen in semen and
male urine. Forensic Sci. Int., 127, 71-74.
Schier, R., McCall, A., Adams, G. P., Marshall, K. W., Merritt, H., Yim, M., Crawford,
R. S., Weiner, L. M., Marks, C. and Marks, J. D. (1996). Isolation of picomolar affinity
anti-c-erbB-2 single-chain Fv by molecular evolution of the complementarity
determining regions in the center of the antibody binding site. J. Mol. Biol., 263, 551567.
Schmitz, U., Versmold, A., Kaufmann, P. and Frank, H. G. (2000). Phage display: a
molecular tool for the generation of antibodies-a review. Placenta, 21, 106-112.
Schulman, C.C., Ekane, S. and Zlotta, A.R. (2001). Nutrition and prostate cancer:
evidence or suspicion? Urol., 58, 318-334.
Shannon, J., Tewoderos, S., Garzotto, M., Beer, T. M., Derenick, R., Palma, A. and
Farris, P. E. (2005). Statins and prostate cancer risk: a case-control study. Am. J.
Epidemiol., 162, 318-325.
Shariat, S., Canto, E., Kattan, M. and Slawin, K. (2004). Beyond prostate-specific
antigen: new serologic biomarkers for improved diagnosis and management of prostate
cancer. Rev. Urol., 6, 58-72.
Simich, J. P., Morris, S. L., Klick, R. L. and Rittenhouse-Diakun, K. (1999). Validation
of the use of a commercially available kit for the identification of prostate specific
antigen (PSA) in semen stains. J. Forensic Sci., 44, 1229-1231.
Smith, G. P. (1985). Filamentous fusion phage: novel expression vectors that display
cloned antigens on the virion surface. Science, 14, 1315-1317.
216
Sokoll, L. J., Chan, D. W., Mikolajczyk, S. D., Rittenhouse, H. G., Evans, C. L.,
Linton, H. J., Mangold, L. A., Mohr, P., Bartsch, G., Klocker, H., Horninger, W. and
Partin, A. W. (2003). Proenzyme PSA for the early detection of prostate cancer in the
2.5-4.0 ng/ml total PSA range: preliminary analysis. Urol., 61, 274-276.
Soukka, T., Paukkunen, J., Harma, H., Lonnberg, S., Lindroos, H. and Lovgren, T.
(2001). Supersensitive time-resolved immunofluorometric assay of free prostatespecific antigen with nanoparticle label technology. Clin. Chem., 47, 1269-1278.
Sozen, S., Eskicorapci, S., Kupeli, B., Irkilata, L., Altinel, M., Ozer, G., Uygur, C.,
Alkibay, T. and Ozen, H. (2005). Complexed prostate specific antigen density is better
than the other PSA derivatives for detection of prostate cancer in men with total PSA
between 2.5 and 20 ng/ml: results of a prospective multicenter study. Eur. Urol., 47,
302-307.
Stangelberger, A., Margreiter, M., Seitz, C. and Djavan, B. (2007). Prostate cancer
screening markers. J. Men’s Health, 4, 233-244.
Stenman, U. H., Leinonen, J., Zhang, W. M. and Finne, P. (1999). Prostate-specific
antigen. Semin. Cancer Biol., 9, 83-93.
Stephan C., Cammann, H., Meyer, H. A., Lein, M. and Jung K. (2007). PSA and new
biomarkers within multivariate models to improve early detection of prostate cancer.
Cancer Letter, 24, 18–29.
Stephan, C., Jung, K., Lein, M., and Diamandis, E.P. (2004). PSA and other tissue
kallikreins for prostate cancer detection. Eur. J. Cancer, 43, 1918-1926.
Steuber, T., Nurmikko, P., Haese, A., Pettersson, K., Graefen, M., Hammerer, P.,
Huland, H. and Lilja, H. (2002). Discrimination of benign from malignant prostatic
disease by selective measurements of single chain, intact free prostate specific antigen.
J. Urol., 168, 1917-1922.
217
Storb, U., Shen, H.M., Michael, N., and Kim, N. (2001). Somatic hypermutation of
immunoglobulin and non-immunoglobulin genes. Philos. Trans. R. Soc. Lond. B. Biol.
Sci., 356, 13-19.
Stowell, L. I., Sharman, L. E. and Hamel, K. (1991). An enzyme-linked immunosorbent
assay (ELISA) for prostate-specific antigen. Forensic Sci. Int., 50, 125-138.
Structural Genomics Consortium (2008). Protein production and purification. Nature
Methods, 5, 135-146.
Struewing, J.P., Hartge, P., Wacholder, S., Baker, S.M., Berlin, M., McAdams, M.,
Timmerman, M. M., Brody, L. C. and Tucker, M. A. (1997). The risk of cancer
associated with specific mutations of BRCA1 and BRCA2 among Ashkenazi Jews. N.
Engl. J. Med., 336, 1401-1408.
Studier, F.W. (2005). Protein production by auto-induction in high density shaking
cultures. Protein Exp. Purif., 41, 207-234.
Thaxton, C. S., Elghanian, R., Thomas, A. D., Stoeva, S. I., Lee, J. S., Smith, N. D.,
Schaeffer, A. J., Klocker, H., Horninger, W., Bartsch, G and Mirkin, C. A. (2009).
Nanoparticle-based bio-barcode assay redefines ―undetectable‖ PSA and biochemical
recurrence after radical prostatectomy. PNAS (USA), 10, 1-6.
Thompson, I. M. Pauler, D. K., Goodman, P. J., Tangen, C. M., Lucia, M. S., Parnes,
H. L., Minasian, L. M., Ford, L. G., Lippman, S. M., Crawford, E. D., Crowley, J. J. and
Coltman, C. A., Jr. (2004). Prevalence of Prostate Cancer among Men with a ProstateSpecific Antigen Level 4.0 ng per Milliliter. N. Engl. J. Med., 350, 2239-2246.
Trojan, L., Bode, C., Weiss, C., Mayer, D., Grobholz, R., Alken, P. and Michel, M. S.
(2006). IGF-II serum levels increase discrimination between benign prostatic
hyperplasia and prostate cancer and improve the predictive value of PSA in clinical
staging. Eur. Urol., 49, 286-292.
218
Ueda, E. K. M., Gout, P. W. and Morganti, L. (2003). Current and prospective
applications of metal ion-protein binding. J. Chromatogr., 988, 1-23.
Varambally, S., Dhanasekaran, S. M., Zhou, M., Barrette, T. R., Kumar-Sinha, C.,
Sanda, M. G., Ghosh, D., Pienta, K. J., Sewalt, R. G., Otte, A. P., Rubin, M. A. and
Chinnaiyan, A. M. (2002). The polycomb group protein EZH2 is involved in
progression of prostate cancer. Nature, 419, 624-629.
Varma, M., Morgan, M., Jasani, B., Tamboli, P. and Amin, M.B. (2002). Polyclonal
Anti-PSA Is More Sensitive but Less Specific Than Monoclonal Anti-PSA Implications
for Diagnostic Prostatic Pathology. Amer. J. Clin. Path., 118, 202-207.
Vihinen, M. (1994). Modeling of prostate specific antigen and human glandular
kallikrein structures. Biochem. Biophys. Res. Commun., 204, 1251-1256.
Weir, H.K., Thun, M.J., Hankey, B. F., Ries, L. A., Howe, H. L., Wingo, P. A., Jemal,
A., Ward, E., Anderson, R. N. and Edwards, B. K. (2003). Annual report to the nation
on the status of cancer, 1975–2000, featuring the uses of surveillance data for cancer
prevention and control. J. Natl. Cancer Inst., 95, 1276-1299.
Whittemore, A. S., Kolonel, L. N., Wu, A. H., John, E. M., Gallagher, R. P., Howe, G.
R., Burch, J. D., Hankin, J., Dreon, D. M., West, D. W. and et al. (1995). Prostate
cancer in relation to diet, physical activity, and body size in blacks, whites, and Asians
in the United States and Canada. J. Natl. Cancer Inst., 87, 652-661.
Wilson, W. D. (2002). Analyzing biomolecular interactions. Science, 295, 2103-2105.
Woodrum, D. L., Brawer, M. K., Partin, A. W., Catalona, W. J. and Southwick, P. C.
(1998). Interpretation of free prostate specific antigen clinical research studies for the
detection of prostate cancer. J. Urol., 159, 5-12.
219
Wu, J.T., Zhang, P., Liu, G.H. and Wilson, L. (1998). Development of an
immunoassay specific for the PSA-ACT complex without the problem of high
background. J. Clin. Lab. Anal., 12, 14-19.
Wu, G., Datar, R. H., Hansen, K. M., Thundat, T., Cote, R. J. and Majumdar, A.
(2001). Bioassay of Prostate Specific Antigen (PSA) using microcantilevers. Nature
Biotechnol., 19, 856-860.
Ye, Z., Tan, M., Wang, G. and Yuan, J. (2004). Preparation, characterization, and timeresolved fluorometric application of silica-coated terbium(III) fluorescent nanoparticles.
Anal. Chem., 76, 513-518.
Ye, Z., Tan, M., Wang, G. and Yuan, J. (2005). Preparation, characterization and
application of fluorescent terbium complex-doped zirconia nanoparticles. J. Fluoresc.,
15, 499-505.
Yokota, M., Mitani, T., Tsujita, H., Kobayashi, T., Higuchi, T., Akane, A. and Nasu,
M. (2001). Evaluation of prostate-specific antigen (PSA) membrane test for forensic
examination of semen. Leg. Med (Tokyo)., 3, 171-176.
Yousef, G. M. and Diamandis, E. P. (2001). The new human tissue kallikrein gene
family: Structure, function, and association to disease. Endocr. Rev., 22, 184-204.
Yu, F., Persson, B., Lofas, S. and Knoll, W. (2004). Surface plasmon fluorescence
immunoassay of free prostate-specific antigen in human plasma at the femtomolar level.
Anal. Chem., 76, 6765-6770.
Yu, H. and Diamandis, E. P. (1995a). Prostate-specific antigen in milk of lactating
women. Clin. Chem., 41, 54-58.
Yu, H. and Diamandis, E. P. (1995b). Prostate-specific antigen immunoreactivity in
amniotic fluid. Clin. Chem., 41, 204-210.
220
Zeder-Lutz, G., Zuber, E., Witz, J. and Van Regenmortel, M. H. V. (1997).
Thermodynamic analysis of antigen-antibody binding using biosensor measurements at
different temperatures. Anal. Biochem., 246, 123-132.
Zhang, W. M., Finne, P., Leinonen, J., Vesalainen, S., Nordling, S. and Stenman, U. H.
(1999). Measurement of the complex between prostate-specific antigen and alpha-1protease inhibitor in serum. Clin. Chem., 45, 814-821.
Zhang, W. M., Leinonen, J., Kalkkinen, N., Dowell, B. and Stenman, U. H. (1995).
Purification and characterization of different molecular forms of prostate-specific
antigen in human seminal fluid. Clin. Chem., 41, 1567-1573.
Zheng, G., Patolsky, F., Cui, Y., Wang, W. U., Lieber, C. M. (2005). Multiplexed
electrical detection of cancer markers with nanowire sensor arrays. Nature Biotechnol.,
23, 1294-1301.
Zhu, Y., Williams, S. and Zwiggelaar, R. (2006). Computer technology in detection and
staging of prostate carcinoma: a review. Med. Image Anal., 10, 178-199.
Zundel , D., Jarry , H., Kestler, D., Holzapfel, G., Bartels, H., Scheit, K. H. and Wuttke
W. (1990). Development and evaluation of an enzyme-linked immunoassay for the
prostate: specific antigen utilizing two monoclonal antibodies. Urol. Res., 18, 327-330.
221